Eli Heckscher, International Trade, and Economic History
Eli Heckscher, International Trade, and Economic History
edited by Ronald Findlay, Rolf G. H. Henriksson, Ha˚kan Lindgren, and Mats Lundahl
The MIT Press Cambridge, Massachusetts London, England
( 2006 Massachusetts Institute of Technology All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher. MIT Press books may be purchased at special quantity discounts for business or sales promotional use. For information, please email
[email protected] or write to Special Sales Department, The MIT Press, 55 Hayward Street, Cambridge, MA 02142. This book was set in Palatino on 3B2 by Asco Typesetters, Hong Kong. Printed and bound in the United States of America. Library of Congress Cataloging-in-Publication Data Eli Heckscher, international trade, and economic history / Ronald Findlay . . . [et al.], editors. p. cm. Includes bibliographical references. ISBN 0-262-06251-8 (hc : alk. paper) 1. International trade. 2. Economic history. 3. Mercantile system. 4. Heckscher-Ohlin principle. 5. Heckscher, Eli F. (Eli Filip), 1879–1952. I. Findlay, Ronald. HF1379.E4 2006 382.01—dc22 10 9 8
7 6 5
2005056285 4 3 2 1
Contents
Preface 1
ix
Introduction
1
Ronald Findlay, Rolf G. H. Henriksson, Ha˚kan Lindgren, and Mats Lundahl 29
I
Theory, Methodology, and History
2
On the Continuing Need for Both the Critical Evaluation of Sources and the Utilization of Economic Theory: Recent Methodological Developments in Economic History 31 Ha˚kan Lindgren
3
The Making of the Economic Historian: Eli F. Heckscher, 1897– 1904 51 Rolf G. H. Henriksson
II
Heckscher-Ohlin Trade Theory
4
Eli Heckscher and the Holy Trinity
89 91
Ronald W. Jones 5
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization 107 Kevin H. O’Rourke
vi
Contents
III Historical Applications of Heckscher-Ohlin Theory 6
Mediterranean Trade in Biblical Times
139
141
Peter Temin 7
Demographic Shocks and the Factor Proportions Model: From the Plague of Justinian to the Black Death 157 Ronald Findlay and Mats Lundahl
8
Explaining World Tariffs, 1870–1938: Stolper-Samuelson, Strategic Tariffs, and State Revenues 199 Jeffrey G. Williamson
IV Mercantilism 9
229
Eli Heckscher and His Mercantilism Today
231
Lars Magnusson 10
Mercantilism: Power and Plenty Through the Lens of Strategic Trade Policy 247 Douglas A. Irwin
11
Mercantilism, the Enlightenment, and the Industrial Revolution 269 Joel Mokyr
12
The Contemporary Relevance of Heckscher’s Mercantilism Deepak Lal 321
V
The Continental Blockade
13
The Continental System After Eighty Years
323
Franc¸ois Crouzet 14
Eli Heckscher, Economic Warfare, Naval Blockades, and The Continental System 347 Lance E. Davis and Stanley L. Engerman
305
Contents
15
vii
The Hanoverian State and the Defeat of the Continental System: A Conversation with Eli Heckscher 373 Patrick O’Brien 407
VI Eli Heckscher and Swedish Economic History 16
Swedish Industrialization 1870–1930 and the Heckscher-Ohlin Theory 409 Lennart Scho¨n
17
Eli F. Heckscher’s Vision of Economic Development
433
Johan So¨derberg 18
From Wartime Provisioning to Barbarous Prosperity: Eli F. Heckscher’s Investigations of Food Consumption in Early Modern Sweden 457 Mats Morell
VII The Man 19
479
A Portrait of Our Grandfather
481
Eva, Einar, Sten, and Ivar Heckscher 20
Eli F. Heckscher Today: A Bibliometric Picture
493
Bo Sandelin 21
When Heckscher Changed Direction: From Social Conservatism to Economic Liberalism 505 Benny Carlson
22
Eli Heckscher on Jewish Assimilation and Zionism Harry Flam
23
Eva Heckscher, 1936–2004 Dag Klackenberg Index
543
541
525
Preface
When the faculty of the Stockholm School of Economics met on November 8, 1928, Eli F. Heckscher, professor of economics and statistics, offered a proposal calling for the establishment of an institute for research in economic history. The faculty, followed one week later by the School’s Board of Trustees, gave its assent, and on June 18, 1929, Heckscher was awarded a personal professorship in economic history and appointed director of the new institute. The research tradition to which Eli F. Heckscher gave his imprimatur still characterizes much of the work performed within the Institute for Research in Economic History (EHF) at the Stockholm School of Economics. This is especially true of the cross-disciplinary nature of the research, but it also applies to the emphasis on business history and historical economics. With the approach of the fiftieth anniversary of Heckscher’s December 23, 1952 death, it seemed only fitting that the EHF Institute should participate in marking the occasion. Thus, a symposium celebrating Heckscher’s many scholarly contributions was organized under the auspices of the Institute. An organizing committee consisting of Ronald Findlay, Rolf G. H. Henriksson, Ha˚kan Lindgren, and Mats Lundahl was given responsibility for the planning and scholarly content of the symposium. The planning was based on our conviction that Eli F. Heckscher, together with Knut Wicksell, Bertil Ohlin, and Gunnar Myrdal, belongs to a group of Swedish economists who have received wide-spread international recognition for their work. At least in Heckscher’s case, his major contributions go beyond economics as such to include the discipline of economic history. Among economists, he is primarily known for his path-breaking work on international trade, while among economic historians his fame rests on his now classic works on Mercantilism, The Continental System, and Swedish Economic History, all of which
x
Preface
have appeared in English translation. In Sweden, Heckscher founded, and indeed molded, economic history as a scholarly discipline. Thanks to his grounding in economic theory and his extensive list of publications, Heckscher remains relevant to this very day. Inevitably, his research results have been modified in numerous regards. Nevertheless, for many modern scholars his body of work remains both a starting point and a challenge. The goal of the symposium, therefore, was to create a forum where economists and economic historians could meet to develop a picture of how relevant Heckscher’s research program and results are perceived to be by current practitioners of the two disciplines. The results of the conference ‘‘Eli F. Heckscher, 1879–1952: A Celebratory Symposium,’’ are presented in this volume. Many individuals have made valuable contributions to the creation of this book. First and foremost there are the authors, who have patiently awaited the results of their labor, as well as the numerous commentators, who are as follows: Lena Andersson-Skog, Bob Coats, Richard Friberg, Ylva Hasselberg, Maths Isacson, Jan Jo¨rnmark, Olle Krantz, Jonas Ljung¨ gren Jan Ottosson, Tom Petersson, Ronny Pettersson, berg, Anders O ¨ rjan Sjo¨berg, Hans Sjo¨gren, Bo So¨dersten, Richard Lars Pa˚lsson-Syll, O Sylla, Kersti Ullenhag, Daniel Waldenstro¨m, and Kurt Wickman. The insights provided by these commentators have without question facilitated the work of preparing the contributed papers for publication. The conference would not have been possible without the support of the SSE and its president Leif Lindmark. Administratively, success, before, during, and after the conference can largely be credited to the self-sacrificing and unselfish work performed by the staff of the EHF Institute, especially Britt-Marie Eisler and Krim Talia. Moreover, without contributions from a number of leading Swedish research funding sources, the conference could not have been held. Thus the organizing committee wishes to extend its heartfelt appreciation for the financial support provided by the Central Bank of Sweden, Jan Wallander’s and Tom Hedelius’s Foundation, Prince Bertil’s Foundation, the Bank of Sweden Tercentenary Foundation, Marianne and Marcus Wallenberg’s Foundation, the Swedish Research Council, and Timbro AB. Last, but certainly not least, a special thank you is due the grandchildren of Eli F. Heckscher: Eva, Einar, Sten, and Ivar. They participated in the project with great enthusiasm, providing family photographs and Eli’s school essay. Their personal recollections of their grandfather,
Preface
xi
which are included in this volume, shed important light on Heckscher as a private person. Our picture of this great scholar would be incomplete without their contributions. Tragically, Eva Heckscher died before she could see her loving account of her grandfather in print. We wish to honor Eva by dedicating this book to her memory, and to thank Dag Klackenberg for the brief sketch of her life that he has provided. New York and Stockholm, 7 June 2005 Ronald Findlay Rolf G. H. Henriksson Ha˚kan Lindgren Mats Lundahl
1
Introduction Ronald Findlay, Rolf G. H. Henriksson, Ha˚kan Lindgren, and Mats Lundahl
The name of Eli Heckscher will forever be associated with two great ideas in economics: the factor proportions theory of comparative advantage and international trade, which bears his name along with that of his distinguished pupil Bertil Ohlin, and mercantilism, that complex and controversial body of thought that he illuminated, as no one has before or since, in his great two-volume treatise of the 1930s. What is perhaps even more remarkable is that most of Heckscher’s relentlessly active scholarly life was spent on neither of these topics but on his unfinished multivolume masterpiece on the economic history of Sweden. Almost incidentally, he also found the time to write another perennial classic, his study of the Continental Blockade, along with numerous other major works in an amazing bibliography of well over a thousand items. It is certain that we will never see a scholar like Heckscher again; among his great contemporaries perhaps only Jacob Viner was comparable in combining theoretical insights with a depth and breadth of knowledge of both economic history and the history of economic thought. In the present volume we attempt to delineate and assess the contributions that this truly extraordinary man made to economic history and economic theory, to the nature of the relationships between these two very different varieties of economic inquiry, and to the controversies over economic policy that he actively engaged in over the course of a scholarly life that spanned two world wars and the Great Depression. Heckscher was both a patriotic Swede and an uncompromisingly cosmopolitan nineteenth-century liberal citizen of the world. To honor his memory, we are fortunate to have been able to assemble a team of Swedish and international scholars whose own research programs have all been inspired by his ideas and example.
2
Introduction
A Biographical Sketch Eli Filip Heckscher was born in 1879.1 His father came from Denmark to Sweden to work with his brother in a banking firm. The two represented the eighth generation of a family of wealthy Jewish bankers who had begun their activities in Hamburg as far back as the sixteenth century. After graduating from high school in 1896, Eli Heckscher began his academic studies in Uppsala the following year, specializing in history. He gravitated to the circle of historians headed by Harald Hja¨rne, a well-known and charismatic figure in Swedish academic life. Hja¨rne was also a major participant in public policy debates, taking a socialconservative stance and strongly advocating social reform. It was in history that Heckscher wrote his licentiat2 thesis in 1903, on the Swedish Navigation Act (Produktplakatet) of 1724 (Heckscher 1908). His studies towards this degree also included economics, taught by David Davidson, and political science. Not surprisingly, Heckscher opted for economic history when it came to selecting a topic for his Ph.D. Before writing his doctoral dissertation, however, Heckscher wrote the first in a long series of papers on methodology, a topic that would concern him for the rest of his life (Henriksson and Lundahl 2003a), especially the relationship between economic history and economic theory. Heckscher argued that the task of economic history is exploring the development of economic life and that in order to do so the economic historian necessarily has to make use of economic theory (Heckscher 1904). During his entire career Heckscher displayed something of a Janus face, being both an economic historian and an economist, and he was forced to come to grips with the similarities and differences between the two subjects, constantly reevaluating and deepening his views.3 Heckscher wrote his doctoral dissertation (Heckscher 1907) in a mere two years. The topic was the economic effects of the development of the Swedish railway system, and the explicit consideration of a counterfactual in some parts of the book makes the work an early precursor of the New Economic History movement (Henriksson 1991: 145). Incidentally, one of the leaders of this intellectual movement, Robert Fogel, also wrote on the railroad’s economic impact in the United States. Heckscher’s thesis earned him the position as docent (the equivalent of an associate professorship) in political economy at Stockholm College in 1907. After two years he was appointed to the chair in
Introduction
3
economics and statistics at the new Stockholm School of Economics, which was founded that same year. Holding this position brought Heckscher increasingly into economics and away from economic history, but his activities during the first decade as a professor were also partly a response to the World War I. This focus led to a monograph on the wartime economy (Heckscher 1915), and he also got involved in the Swedish War Commission, first as secretary and then as its executive head. Heckscher’s activities for the commission provided a foundation for his extensive work on this topic in the 1920s for the Carnegie Foundation (Bergendal et al. 1930), and also proved fruitful in other ways. The World War I experience had an obvious historical parallel in the Napoleonic Wars, and this similarity led Heckscher to begin his study of the Continental Blockade System in 1916. It was completed in 1918, and four years later an English edition appeared (Heckscher 1922). In 1919, Heckscher published the work for which he is renowned, not only by economic historians but also by economists, the seminal paper on foreign trade and the distribution of income that provided the foundation for the Heckscher-Ohlin (or factor proportions) theory of foreign trade. This paper was one of only four that by and large comprise Heckscher’s entire theoretical output, all of which were published within a narrow time span between 1916 and 1924.4 Thereafter, Heckscher left his theoretical work behind and set out to fulfill what he considered to be his ‘‘real’’ task in life: writing an economic history of Sweden. The work was delayed, however, because before embarking on this major undertaking he had to complete a few ‘‘minor’’ research tasks. One of these tasks, by no means a minor one, was his theoretical history of mercantilism. The monumental study was published in 1931 and translated into English four years later (Heckscher 1935a). It immediately became a standard work and was hailed as a classic. In 1931 Heckscher also published his survey of the Industrial Revolution (Heckscher 1931), a work that became a long-lived textbook in Swedish universities, appearing in several editions, but that has unfortunately never been translated into English. A second task that Heckscher could not refuse was active participation in the formation of Swedish economic policy. After World War I he became a member of the Commission on Tariffs and Treaties, whose final work very much bore the imprint of Heckscher’s free trade views, and in 1926 he was a member of an unemployment committee. Throughout his life Heckscher
4
Introduction
considered it more or less a duty to contribute to government policy making and to the public debate of economic issues in general. The (incomplete) bibliography of his published output (Eli Heckschers bibliografi 1950) lists no less than 1,148 works, most of which are newspaper articles and similar projects. Altogether the 1910s and (to a lesser extent) the 1920s were the decades of Eli Heckscher the policy-making economist, as distinct from the economic historian. Heckscher held the economics and statistics chair at the Stockholm School of Economics until 1929, and he was quite influential when it came to the organization of teaching there, striking a balance between the curriculum’s scientific foundations and his endeavor to provide the young men who would later make careers mainly in private business with a grounding in economics that allowed them to deal efficiently with practical problems. He attempted to uphold high standards, and was widely feared as an examiner. Only a handful of the approximately one thousand students that he examined obtained the highest grade, one of them Bertil Ohlin. A lively description of Eli Heckscher the teacher during this period was provided in the mid-1960s by Ruben Rausing, the founder of the Swedish corporation Tetra Pak. Rausing had entered the Stockholm School of Economics in 1916 and, like Ohlin, had become part of the select group of Heckscher’s best students: He greatly influenced my life. It is also of him that I have my richest and warmest mental image of those years. When heading for the lectures he never walked. Instead he ran and jumped nimbly up. In formal instruction, he was an extraordinary lecturer, lively, clear and fully engaged with his students. This contact was even further intensified during seminars. There could be sharp exchanges between professor and student. He combined this ability to relate to the students with a remarkable and unpretentious dignity. He himself was always punctual to the minute. If a student arrived late, he would pause meaningfully. The person in question would learn that punctuality is a requirement of all organized work and of all civilization.5
Heckscher’s involvement with the Stockholm School of Economics changed in 1929, after an incident that ended on a sour note. When the president of the school, Carl Hallendorff, unexpectedly died that year, Heckscher regarded himself as the natural successor, but his colleagues did not, especially those who saw him as condescending, sarcastic, and arrogant. The result was that Heckscher found himself without support and considered leaving the school. The problem was, however, happily solved by the creation of a special chair in economic history
Introduction
5
under the auspices of the new Economic History Institute. This position freed Heckscher from teaching duties and allowed him to concentrate entirely on research. With this new appointment, Heckscher the economist took the opportunity to finally become Heckscher the economic historian. He chose to leave theoretical work and contemporary economic issues aside, except for when exceptional circumstances called for his presence in ongoing policy debates, and to concentrate on what he considered his real mission in life. Without the newly established chair, which Heckscher occupied until reaching retirement in 1944 and then by special extension to 1949, it would probably not have been possible for him to write his projected four-volume magnum opus, Sveriges ekonomiska historia fra˚n Gustav Vasa (The Economic History of Sweden since Gustav Vasa). Heckscher had begun this work in the early 1920s, but it was not until he could concentrate more or less exclusively on research that he could fully master the material necessary for his monumental survey. The first installment, on the sixteenth century, was ready in 1935 (Heckscher 1935b), and the volume on the seventeenth century followed duly in 1936 (Heckscher 1936), but the third part on the eighteenth century did not appear until 1949 (Heckscher 1949a, 1949b). The reason was simple. The amount of work involved had increased as Heckscher’s study advanced chronologically in time. The material grew richer and the statistical series longer. Heckscher, whose research increasingly was based on the painstaking accumulation and analysis of statistics, ended up overtaxing his health, and by the time the twopart eighteenth-century volume was completed the notorious workaholic (who had had two collapses in 1942 and 1943) was a physically broken man who only had a few years left to live. The last volume, on the nineteenth century, remained unwritten. The approach that Heckscher used, basing his analysis on the close examination of voluminous statistical series, would simply have been too onerous to allow him to complete the work.6 Eli Heckscher died in December of 1952. Among economists he is known mainly for his seminal contribution to international trade theory. Had he been alive in 1977 he would no doubt have shared the Nobel Prize in economics with Bertil Ohlin and James Meade. Among economic historians Heckscher’s reputation rests mainly on his monumental works exploring mercantilism, the Continental Blockade, and Sweden’s economic history. As we have already pointed out, he
6
Introduction
presented to the scholarly world something of an intellectual Janus face. Yet during his entire scientific life Heckscher struggled with the question of how to bridge the divide between economics and economic history. The purpose of the present work, which emanates from the fiftieth anniversary of his death, is to provide readers with a complete portrait of Heckscher’s scientific contributions, both in economics and in economic history, and furthermore to link his distinctive approach to later developments in both areas. The book also gives a portrait of Heckscher the private person. Theory, Methodology, and History Ha˚kan Lindgren starts off his discussion of methodological issues in chapter 2 with the research program for the new field of economic history that Eli F. Heckscher, at the tender age of twenty-four, published in 1904. In this programmatic paper, the German historical school is criticized for either getting lost in a sea of details or else trying to squeeze history into some ill-fitting stage theory of economic development. On the other methodological flank, Heckscher argued that deductive economics suffers from a tendency to view economic laws as universal laws of nature, not as historical categories. It was his contention that a new field, economic history, could be based on a melding of the best features of both these parent disciplines. During recent decades, the methodologies employed in economics and history have moved even further apart. History has become receptive to micro history and to social constructivism, while economics has become ever more deductive and mathematical. Given these divergent methodological trends, Lindgren poses the question of whether the young Heckscher’s hope of merging history and theoretical and empirical economic inquiry still has relevance, or if his proposal was stillborn. A review of the Swedish experience leads to the conclusion that Heckscher’s 1904 program still remains relevant today. Both the historian’s evaluation of sources and the economist’s models are needed to resolve problems in economic history. Social constructivism, in the sense that there are ‘‘cultural codes’’ that must be deciphered, is an essential element in all historical research, while economic models have great explanatory power when applied to appropriate economic historical questions. As Heckscher argued, however, economic history is not historical economics and should not evolve into a branch of economics.
Introduction
7
Economic history is an empirical, inductive discipline, but by no means is it nontheoretical. Knowledge of modern ‘‘pure theory,’’ as well as of previously accepted theories, can be very helpful in explaining how economies have been organized and how these systems functioned during different historical eras and in different cultural settings. Lindgren therefore argues that the emphasis placed on economic theory and econometric methods in the current economic history curriculum should be increased, furthering the agenda Heckscher articulated 100 years ago. In chapter 3 Rolf Henriksson elaborates on the methodological discussion initiated in the previous chapter. Heckscher’s position, occupying the middle ground between history and economics, as presented in his 1904 manifesto, is traced back to the formative years of his academic career. When the eighteen-year old Heckscher matriculated at Uppsala University in 1897, history became his major field, and he was admitted to Professor Harald Hja¨rne’s inner circle of top students. Hja¨rne was an avid supporter of Leopold von Ranke’s program for historical methodology and its focus on historical understanding, striving for objectivity and the critical evaluation of sources. In the course of his economics studies, Heckscher came into contact with Professor David Davidson, who held the economics and fiscal law chair on the law faculty. Davidson had been influenced by the historical school and by socialist economic thinking. Although he was a great admirer of David Ricardo, he was not receptive to the ongoing neoclassical revolution in economics. At that time, the great majority of economics students were candidates for a law degree. Thus, when Heckscher chose economics as the second field in his licentiat degree program, he became one of a select few students who seriously pursued graduate studies in this discipline. Moreover, during the last few years of his university studies, Heckscher worked very closely with Davidson as an unofficial assistant. On the basis of Heckscher’s program of university studies, as well as on his early writings (including his licentiat thesis in history), Henriksson identifies and traces three different lines of influence that would become the pillars of the 1904 manifesto. Having been trained in the Hja¨rne tradition, reliance upon and critical evaluation of primary sources were basic principles of Heckscher’s scholarly program. Indeed, its importance to him only increased over time. His plea for deductive economic analysis and his conceptualizing approach to the social sciences can be traced back to Davidson’s influence. As for
8
Introduction
Heckscher’s advocacy of the use of statistics, this can be related to his study of political science, even though, like Hja¨rne, he remained skeptical of the official variety. The originality in Heckscher’s 1904 manifesto lay in his having selected some, and rejected other, aspects of these three traditions and then forging his choices into a whole new program. Heckscher-Ohlin Trade Theory The second section of the present volume focuses on Eli Heckscher the trade theorist. Heckscher’s 1919 paper on international trade was not translated into English until thirty years later, and then only partially. (It was not until the appearance of the 1991 volume edited by Harry Flam and June Flanders [Heckscher and Ohlin 1991] that a full translation of the original article became available.) Before this, scholars without a reading knowledge of Swedish were unable to evaluate the extent and originality of Heckscher’s own contribution to the model that bore his name jointly with that of Bertil Ohlin. In chapter 4, however, Ronald Jones deconstructs the 1919 paper, pointing out exactly what Heckscher did and did not do in that celebrated piece. He notes that Heckscher clearly stated the possibility that free trade between economies endowed with different factor proportions would lead to a complete equalization of factor prices and would therefore be a perfect substitute for international factor mobility. Jones also observes that Heckscher, along with Ohlin, did not state some of the crucial assumptions necessary for the factor-price equalization theorem to hold: that the number of factors could not exceed the number of goods that are traded, and that the possibility of factor-intensity reversals has to be excluded. Yet in 1919 nobody had the technique necessary to appreciate these aspects of the complex mathematical structure underlying the theorem’s proof, which only became known after the work of Paul Samuelson in the late 1940s and early 1950s. Indeed, a completely rigorous proof did not exist until after the work of the mathematicians David Gale and Hukukane Nikaido (1965) on the global univalence theorem. That Heckscher, an economic historian by temperament and training, was single-handedly able to first envision and grasp the underlying economic logic of this deep and beautiful theoretical result is nothing short of astonishing. In chapter 5 we turn from the logical abstractions of the factor-price equalization theorem and related theoretical issues of the HeckscherOhlin model to some very basic empirical tests of the predictions that
Introduction
9
the model makes about individual attitudes across countries toward globalization, or more specifically toward national policies on trade and immigration. Kevin O’Rourke begins with the simplest version of the model, with skilled and unskilled labor as the two factors. Since trade favors the abundant factor and harms the scarce one, skilled workers should be against protection in richer, more skill-abundant countries and in favor of it in poorer, less skill-abundant countries. Similarly, skilled workers in richer countries should favor the immigration of unskilled workers, who are complementary to them, but skilled workers in poorer countries should be against immigration of skilled workers, who are competitive with them. Unskilled workers in each type of country should of course have the opposite preferences to those of their skilled compatriots. O’Rourke tests these predictions rigorously on cross-country data from a sample of twenty-four countries and generally finds them to hold remarkably well, with the results also supported by independent studies undertaken by Matthew Slaughter, Dani Rodrik, and others. As O’Rourke points out, these results refute the plausible hypothesis that more educated skilled workers everywhere support globalization because they are more ‘‘enlightened’’ than their less educated unskilled fellows, who are everywhere against it. It is thus not a question of what type of worker one is, regardless of location, but whether one is the abundant or the scarce factor in a given location, that seems to matter for attitudes toward globalization. This conclusion is just as the model predicts. Historical Applications of Heckscher-Ohlin Theory The third section deals with the application of Heckscher-Ohlin trade theory to historical material. Strangely enough, Heckscher himself never attempted to apply his seminal 1919 paper to the analysis of historical material (Findlay and Lundahl 2002). His use of trade theory was limited to the discussion of current economic problems (Henriksson and Lundahl 2003a). Nevertheless, as the three chapters by Peter Temin, Ronald Findlay and Mats Lundahl and Jeffrey Williamson demonstrate, the factor proportions approach provides a powerful tool in a wide variety of historical settings and episodes, across a time period of over two millennia. In chapter 6 Peter Temin deals with trade in the Mediterranean during biblical times. Temin bases his discussion of the existence of trade and the reasons for it on archaeological evidence: two sunken ships
10
Introduction
from the latter half of the eighth century B.C., found north of the Egyptian coast, transporting amphorae containing wine from Phoenicia to Egypt, obviously to be traded there. Archaeological evidence for early trade in the Mediterranean also comes from a thirteenth-century B.C. wreck off the coast of Turkey and from a painting in a Theban tomb of Syrian ships arriving in Egypt. Thus, interregional trade existed in the Mediterranean during the Bronze Age. This trade, however, appears to have been broken off during the transition to the Iron Age in the twelfth century B.C. but to have been revived in the eighth, and there is further evidence of trade in the fifth century B.C. Temin deals with trade in bulky commodities—like grain, oil, and wine—commodities that required a lot of coordination between different agents in the commercialization chain. He argues that this coordination was achieved through the market mechanism, with unminted silver being used as the medium of exchange. Temin also attempts an explanation of the trade pattern along Heckscher-Ohlin lines. Egypt exported wheat and Phoenicia exported wine. According to Temin, fertile land was a scarce factor in Egypt, its extension being determined by the flooding of the Nile, so wheat ought to have been the laborintensive commodity, while wine was land-intensive, unless trade was determined by climatic factors alone. Finally, Temin deals with the risks of trade, the high transaction costs these entailed, and the attempts to bring the costs down. The three main risks were natural disaster on the voyage, piracy, and nonpayment or faulty payment. The first could be managed by not venturing too far on to the high seas. Piracy required arms and/or escorts, and payment, in the absence of international treaties, had to be ensured by a reputation equilibrium signaling fair deals. Temin argues that piracy must have been responsible for much of the break-off of trade in the twelfth century B.C., and that as trade revived in the eighth century B.C. deliberate efforts were made to reduce trade risks. Chapter 7, by Ronald Findlay and Mats Lundahl, moves the time perspective to the eight centuries that elapsed between the Plague of Justinian and the Black Death, that is, from the sixth to the fourteenth centuries A.D. The issue is how the advent of the earlier plague affected three civilizations or economic regions and the extent to which each of these was able to recover from or perhaps even transcend its effects: the Byzantine Empire, where the plague struck directly and where the economy barely recovered centuries later; the Islamic world, which fared much better and even experienced a green revolution
Introduction
11
based on crops introduced from the east; and Western Europe, which initially was the most economically backward region but eventually expanded most rapidly of all on the basis of land clearance and the use of the heavy plow and other agricultural innovations. Findlay and Lundahl develop a factor proportions–based model that produces some key predictions regarding the impact of demographic change and can be used to shed some light on the difference in performance between the three geographical regions. The model rests on a Malthusian specification of birth and death rates in combination with an endogenous land frontier, the extension of which is subject to increasing marginal costs. It produces a steady state determining population and labor force size, per capita consumption, total output, and the land area under cultivation, as well as the rewards to land and labor. The model can be subjected to different kinds of exogenous shocks, notably a reduction of the population but also, for instance, technological change in production. Findlay and Lundahl show that when the population shrinks, wages increase and land rents decline. The amount of new land cleared is reduced and so is the area under cultivation. Per capita output and consumption rise and trigger new population growth. Eventually the price of land rises and the real wage declines. New land is added and the economy returns to the steady state. It is also demonstrated that technological change will increase both the population and the area under cultivation. The model can be used to shed some light on the difference in performance between the three geographical regions. In the Byzantine Empire the plague caused the economy to contract and the agricultural frontier to retreat, but higher per capita income and wages made the population grow back, altering the labor market balance in favor of landowners. Ultimately, the absence of technological progress and the deterioration of political and military conditions made the empire lose territory and collapse. The Islamic world, on the other hand, filled the vacuum created by the plague in the eastern Mediterranean. At the outset of the period it enjoyed territorial expansion, but by conquest rather than land clearance. This expansion was accompanied by technological progress in agriculture and population growth. However, the latter was constrained by the region’s harsh geographical conditions and by the Mongol conquests toward the end of the period. Western Europe, finally, was the area that fared best. There the ravages of the plague were not as harsh as in regions further east, and the recovery was accompanied by technological progress on the one hand,
12
Introduction
and eastward territorial expansion by conquest on the other. Altogether these developments made for, comparatively speaking, spectacular population growth, to the point of overpopulation shortly before the Black Death arrived in Western Europe. With chapter 8 we move four to five hundred years further ahead in time. Jeffrey Williamson’s essay deals with the determinants of tariff policy in thirty-five countries from 1870 to 1938. During this period two great surges in tariff levels took place, the first from 1865 to about 1900, and the second in the 1920s and 1930s, with great geographical variations both between and within regions. Latin America, the European overseas offshoots and the European periphery led the pre-1914 backlash, while in the second surge tariff levels rose everywhere. Williamson’s chapter concentrates on a discussion of the mechanisms that may have caused the tariff increases. Four possible explanations are introduced: Stolper-Samuelson tariffs driven by the scarce factors in each region (the factors that stood to gain from tariffs), infant industry tariffs (intended to foster industrialization), revenue tariffs, and, finally, strategic trade policy tariffs (tariff increases triggered by high tariffs among the trading partners). The infant industry argument is ruled out because high tariffs were correlated with slow, not rapid, growth, and hence would not have caused these increases. Williamson proceeds to test the remaining three explanations econometrically, controlling for the effect of inflation on ad valorem tariff rates (during a period when the actual duties were specific) and other factors affecting these rates. The results of the time series regressions indicate that all three explanations played a part. The most important determinant of tariff increases for the entire 1870–1938 period turns out to be increases in the tariff levels of the trading partners. However, as demonstrated by other recent contributions, before 1900 the picture was more divided. Fears of deindustrialization in the periphery and cheap grain imports in the core both played a major role. Political economy effects resulting from increased urbanization and increased skill levels also had an impact. Partner tariffs were more significant in the European periphery, compensation for transport cost reductions influenced events in the industrialized European core, in Asia, and in the non-Latin American overseas offshoots of Europe. The reduction of tariff levels from the 1890s to the outbreak of World War I was caused by rising income levels in Europe and its overseas offshoots, including Latin America, as a result of mass migration and capital movements.
Introduction
13
Finally, the rising tariff levels during the interwar period were driven almost exclusively by strategic motives. The cross-section regression results differ from the time series data by not conceding any importance to partner tariffs. Tariff levels (as opposed to changes) were not determined by partner levels but by other considerations. For example, the low initial tariff levels in Asian countries were the result of their colonial status vis-a`-vis trading partners with high tariff levels; that is, these were forced on the colonies by their European rulers. Mercantilism Heckscher’s Mercantilism is the first modern work on the topic. As such it has been the natural starting point for subsequent research in the field. The study was begun during his years in Uppsala but was not published until 1931. The focus is on economic policy rather than on actual historical events, and Heckscher paints a panoramic picture extending from the Middle Ages until the age of liberalism in the nineteenth century. The work benefited from employing hitherto unused sources, but it was also informed by theoretical knowledge. To a very large extent Mercantilism is a work of synthesis, (Montgomery 1953: 168), a consideration that constantly occupied Heckscher (Olsson 1992). In different ways all the four chapters in the fourth section relate modern research on mercantilism to the foundation provided by Heckscher. In chapter 9 Lars Magnusson traces the evaluation of Mercantilism over time by using an explicitly historical strategy: exploring how the interpretation of mercantilism has changed over time, and how these changes have been influenced by the context of each period. He rejects attempts to reconcile mercantilist ideas with modern theory because this approach anachronistically uses the wrong yardstick by failing to do justice to the early theories. When Mercantilism was published it was criticized on the basis that the relationship between economic ideas and political action was tenuous, and that Heckscher’s attempt to come up with a synthesis melding the two did not hold water. Magnusson reconstructs Heckscher’s general argument that the mercantilists had a set of ideas. The goal of the mercantile system was to maximize state power, not wealth: the core of the system was the means employed to reach the goal, and the state was the propelling agent of the system. This view was criticized
14
Introduction
by Jacob Viner, who argued in favor of the approach that modern theory takes as its point of departure: the mercantilists confused money and wealth, they were bullionists to the core, and they were the representatives of a policy built on special interests. British economists during the 1950s, 1960s, and 1970s regarded Heckscher’s conception of mercantilism as too broad, as it simply gave too much unity to what were in reality disparate events. The notion of a mercantile state was rejected as a straw man. The basis of mercantilist ideas was sought in (sometimes very specific) contemporary events of the seventeenth and eighteenth centuries. The obsession with bullion was seen as the outcome of bilateral trade (Heckscher denied that bilateral trade was common). Magnusson concludes by providing a current perspective on Mercantilism, asserting that many of these criticisms remain valid. Economic ideas and policy practice are not as closely related as Heckscher argued, and it is hardly possible to draw any clear demarcation line between ‘‘mercantilist’’ and ‘‘liberal’’ ideas. Still, Magnusson argues, Mercantilism is worth reading. There was definitely a ‘‘mercantilist’’ economic worldview, and mercantilism displays at least some of the system-like characteristics emphasized by Heckscher. His impressive knowledge of ideas and policies still has a lot to offer. Finally, even though he might overstate his case, Heckscher’s insistence on the nexus between ideas and policies cannot be neglected. Although the ability of the early modern states to carry out coherent economic policies left a lot to be desired, policymakers were to a certain extent influenced by ideas. Chapter 10, by Douglas Irwin, takes as its point of departure the discussion between Heckscher and Jacob Viner of whether mercantilism was mainly about power (Heckscher) or plenty (Viner). With the aid of the modern theory of strategic trade behavior, Irwin analyzes the competition between the English and Dutch East India companies during the first decades of the seventeenth century and demonstrates that there need not be any conflict at all between the two goals. Power and wealth may be complementary rather than competing ends. Irwin begins with a brief exposition of strategic trade policy in a Cournot duopoly model and shows how once a Nash equilibrium has been established between the two firms, government export subsidies may shift this equilibrium in favor of the domestic firm. This finding provides the theoretical foundation for the rest of the chapter. Next, Irwin recapitulates some mercantilist economic ideas: the desire to ex-
Introduction
15
port is linked to government export promotion, establishing the relevance of strategic trade policy. The following section is devoted to a reconciliation of the empirical facts of the East India trade with the strategic trade policy model, and to a comparison of Dutch and British policy. Irwin argues that the English East India Company was simply maximizing profits, while its Dutch competitor maximized a linear combination of profits and revenue. The difference arose from the fact that the English company was a purely private firm seeking profits while the Dutch company’s policy was determined by the so-called bewindhebbers, who received both dividends and a provision based on turnover. The creation of the state-supported Dutch East India Company made it possible for the Dutch to increase both sales volume and profits and made the Dutch company a Stackelberg leader. Chapter 11, by Joel Mokyr, deals with the ‘‘end’’ of mercantilism as ushered in by the Enlightenment and the Industrial Revolution. Heckscher identified intellectual changes as one of the roots of the latter, and one of the main strands of the Enlightenment was the stress on economic progress. The Enlightenment also brought institutional changes of rules, norms, and beliefs, to a large extent because this period’s leading actors were favorably disposed to innovations. Mokyr concentrates on these institutional changes. Rent seeking was an important part of mercantilism. The European economies were caught in a high rent seeking equilibrium with low returns to productive efforts. L’ancien re´gime to a large extent was a machine that redistributed wealth and income. The French Revolution put an end to this, and in the wake of revolution a number of deregulations and eliminations of entry barriers followed. These went hand in hand with a change in the dominant ideology guiding rulers and civil servants. The attitude toward protectionism changed. Whether this was the result of Enlightenment ideas is, however, debatable, especially since contemporary writers were not in agreement on the subject, and in Prussia the concept of Nationalo¨konomie eventually superseded free trade. The Enlightenment also encouraged inventions, notably through the patent system, although some thinkers cautioned against the potentially destructive forces inherent in technological change. Mokyr concludes by stressing the role of institutional progress in the Industrial Revolution. If institutions had not changed, the fruits of technological change might easily have ended up in the hands of rent seekers, tax collectors, and the like, and this would have slowed the pace of growth. It was the combination of ideological and institutional
16
Introduction
change on one hand and the growth of knowledge on the other that produced and fed the Industrial Revolution. What produced the Enlightenment is a more difficult question to answer, but clearly scepticism and rejection of old ideas had a great deal to do with it, and the new ideas went hand in hand with the emergence of free markets and international trade. Deepak Lal, in chapter 12, applies one of the findings of Heckscher’s Mercantilism to today’s international economy. The phenomenon that he investigates is the cycle of economic repression and liberal reform. Lal’s starting point is the creation of dirigiste states, eagerly desired by rulers intent on nation building, in most third world countries during the 1960s and 1970s, followed by liberal reforms from the 1980s onward in the face of combined fiscal and balance-of-payments crises. As Eli Heckscher argued in Mercantilism, dirigisme was a pattern that characterized the rise of the European states during the Renaissance but that fostered corruption, tax evasion, rent seeking, and illegal economic activities. These problems led to an erosion of the tax base during the eighteenth century, before reforms set in during the nineteenth century. The ‘‘crisis theory of economic liberalization’’ is thus applicable to both historical episodes. It is not until the state begins to wither away and society threatens to break down in a crisis situation that wholehearted reform is triggered. Lal also applies the crisis theory to the question why dirigiste policies made a comeback during the twentieth century. To this end he adopts the distinction between the state as a civil association that does not impose any preferred pattern of goals and the state as an enterprise association that uses the law for its own purposes. Classic liberalism embraces the former viewpoint and socialism the latter one. The dissolution of the ‘‘morality of communal ties’’ inherited from the Middle Ages produced individuality, but it also bred an ‘‘anti-individual’’ who, placing a high premium on security, preferred not to make his or her own choices but left these to the government. This collectivist morality tipped the balance in favor of the enterprise view of the state and produced a new wave of dirigisme. Finally, Lal attempts to answer the question of whether the present age of reform is likely to be more long-lived than its nineteenth-century counterpart. The worldwide growth in tax resistance and the virtually complete integration of financial markets plus the recent moves of China and India from planned to market economies all tend to work in favor of reform. However, the social democratic political agenda, the
Introduction
17
realization of the European Union, the fear in the West of economic competition from the East, the increasing substitution of high-skill jobs for low-skill ones in Western countries, and the financial crises in Asia all point in the direction of less globalization, more protectionism, and more dirigisme. Possibly, a third Heckscherian cycle will emerge. The Continental Blockade Heckscher’s study The Continental System grew out of his research on the economic aspects of the 1914–18 World War, when Sweden and most of continental Europe were cut off from overseas trade as a result of the hostilities. From the outset he characterized the Contintental System as Napolean’s gigantic exercise in mercantilist economic policy, aimed at defeating the ‘‘nation of shopkeepers’’ by denying them their export markets and not, as might seem more rational, by depriving them of imports of food and other necessities. Needless to say, the liberal cosmopolitan Anglophile Heckscher did not look kindly on the economic policies of the interventionist emperor in whose mind ‘‘France was first and foremost.’’ Heckscher claims that the blockade failed to cripple Britain economically, but also that this failure was inevitable. Furthermore, the stimulation to French industry was artificial and short-lived despite successes in some sectors such as cotton textiles, sugar beets, and chemicals. Heckscher concluded that the ‘‘selfish’’ way in which Napoleon enforced the blockade on his allies in Europe was severely detrimental to their interests. In chapter 13 Francois Crouzet provides a warm and generous, but not uncritical, appraisal of a work that he calls a ‘‘masterpiece.’’ He agrees with Heckscher that the attempt to strangle Britain by depriving her of export markets was in the end a failure, but he vigorously denies that this outcome was necessary or inevitable and not contingent on circumstances. He points to the severity of the economic crisis in Britain from 1810 to 1812, when exports fell to two-thirds of their 1809 level. Who knows what would have happened if Napoleon’s Russian campaign had not ended in disaster before Moscow in 1812? Crouzet approvingly quotes his ‘‘old master,’’ Georges Lefebvre, to the effect that ‘‘it was not the laws of economics but the Russian winter’’ that caused the blockade to fail. Crouzet finds Heckscher guilty of anachronism in several of his judgments on Napoleon’s actions and perhaps overly critical in his assessment of the blockade’s effects on France.
18
Introduction
The appraisal of the Continental System continues in chapter 14, written by Lance Davis and Stanley Engerman, who embed their narrative and appraisal of this episode’s main events within a broader survey of the history of naval blockades and economic warfare. They extend the discussion beyond Europe to consider relevant events in the New World, such as the slave revolt in Haiti, the War of 1812, and British trade with Latin America, which was stimulated by the restriction of exports to Europe. After an extensive examination of the more recent literature, they conclude that Heckscher seems to have anticipated most of its main results and that his work survives essentially unscathed. They also observe, ironically, that it is Heckscher’s much more controversial work on mercantilism that has continued to keep his name before the scholarly public outside Sweden rather than The Continental System, his less ambitious but more definitive study. Any reader convinced by Davis and Engerman that Heckscher’s analysis of the Continental Blockade is no longer controversial will be startled to read chapter 15 by Patrick O’Brien. Casting his paper as a ‘‘conversation’’ with Heckscher, O’Brien challenges the master headon. Whereas Heckscher frames his account of the blockade as a defeat of Napoleon’s irrational and misguided mercantilist ideology by the pure light of liberal Smithian reason that guided Britain, O’Brien places it in a much broader context: the sustained geopolitical rivalry between France and Britain for power and profit from 1688 to 1815 in which high and rising levels of investment in the Royal Navy gave Britain mastery of the seas and global trade despite France’s continued military strength on the European mainland. Britain was able to do this because the superiority of her fiscal and financial system sustained a balance of overall military power in her favor, with subsidies to allies on the mainland complementing her own command of the sea. In O’Brien’s opinion the blockade was a desperate and doomed attempt to strike at Britain and cause her to accede to French domination of Europe. The plan was to force Britain to lose specie through a negative trade balance by shutting off her access of European export markets, thus undermining the linchpin of her monetary and fiscal system. Even though the policy failed, it should not be condemned as irrational since it was the only option available to France in a protracted geopolitical struggle. O’Brien criticizes Heckscher for two failures of historical objectivity: his Anglocentrism and what O’Brien regards as an excessive attachment to the tenets of classical economics that prevented
Introduction
19
him from appreciating the practical difficulties and wartime exigencies that economic policy makers on both sides had to contend with. It was only after her great continental rival had been decisively defeated by 1815 and her own global hegemony safely assured that Britain could turn to the implementation of the liberal world economic order based on free trade and the gold standard. This stability would be sustained until 1914 and the emergence of another continental challenger. Eli Heckscher and Swedish Economic History In the sixth section of the book some of Heckscher’s basic concepts and interpretations concerning the Swedish economy and its history are reevaluated in light of recent research in the field. Lennart Scho¨n’s contribution in chapter 16 discusses the explanatory value of the Heckscher-Ohlin theorem in relation to the process of Swedish industrialization from 1890 to 1930. Scho¨n makes a distinction between two different types of forces that drove this process: an external, or exogenous, force based on international integration and specialization and an internal, or endogenous, force based on the ability of the domestic economy to interact with new market opportunities and technological innovations. He makes this distinction operational by classifying Swedish industries as consisting either of Heckscher-Ohlin firms, exploiting abundant factors of production, or Schumpeterian firms, exploiting new knowledge and innovations. For Heckscher and Ohlin, international trade was the principal force behind the integration of and factor-price equalization among different national economies. Trade and other open-economy forces worked to increase the price of relatively abundant factors and to reduce the price of those that were relatively scarce. Scho¨n finds that the importance of the two types of industries changed over time. Until the 1890s, Swedish exports were based mainly on abundant raw materials and cheap labor (‘‘Heckscher-Ohlin’’ industries). When changing relative factor prices eroded the advantage originally held by these industries, the Swedish industrial structure responded. During the so-called Second Industrial Revolution, electricity-utilizing and knowledge-intensive industries (‘‘Schumpeterian’’ industries) took the lead. The convergence of factor prices that had characterized the earlier period, both internationally and within Sweden, slowed down and ceased. Indeed, starting in the 1890s and accelerating during World War I, it was replaced by factor price divergence.
20
Introduction
In chapter 17 Johan So¨derberg places Heckscher within the research tradition that has dealt with the overarching question of the long run evolution of economy and society. Here can be found almost all of the great classical economists of the eighteenth and nineteenth centuries; Adam Smith, David Ricardo, John Stuart Mill, and Karl Marx, as well as original thinkers of the early twentieth century such as Thorstein Veblen and Joseph Schumpeter. Eli Heckscher also devoted considerable thought to more ‘‘existential’’ questions, such as the eventual consequences of economic growth and the general forces, both economic and noneconomic, underlying growth. Despite the fact that Heckscher never endowed his model with a formal theoretical structure, So¨derberg concludes that he maintained a vision of the process of long-term economic and societal growth. Economic activity—in Heckscher’s terminology, all human behavior intended to deal with the problem of scarcity—had a single purpose: to make life as rich and worthwhile as possible (that is, to produce economic welfare in the broadest sense). According to Heckscher’s universal interpretation of long-run social development, preferences are endogenous. As wealth increases, the propensity to consume decreases and consumer preferences shift in the direction of nonessential goods and services, thus destabilizing core demand. The role of economic factors in long-run change was limited, especially before the nineteenth century. Noneconomic factors, such as perceptions and attitudes, were the principal sources of economic growth. As to the ultimate outcome of long-term economic development, Heckscher’s expectations for the twentieth century were distinctly gloomy. In this regard, he might be perceived as a prophet of the dismal science along the lines of Ricardo and Malthus. His gloom, however, was his own. Heckscher believed that progress was possible, but by no means was it an inevitable feature of the economic system. The liberal spirit that characterized the nineteenth century had made it an epoch of not only monumental economic growth, but also one of extraordinary cultural progress. The experience of the interwar period, however, did not bode well for the rest of the twentieth century. Changing consumer preferences made the economy less stable, and the market economy, the true source of wealth and prosperity, proved unable to resist the inroads of the state. Individual freedom was being restricted, and the individual was doomed to become the serf of an ever more oppressive state. Clearly, Heckscher’s forebodings were not unlike those of Schumpeter and Hayek.
Introduction
21
A number of Heckscher’s conclusions concerning Swedish economic history have been questioned and revised by subsequent scholarship. Prominent among these is his series of estimated food consumption in Sweden for the extraordinarily long period 1550 to 1815, which Heckscher also argued could be used as a proxy for the overall standard of living. Immediately following its publication as part of the first volume of Sveriges ekonomiska historia, Heckscher’s analysis of Swedish food consumption in the sixteenth century spurred an intense debate among Swedish scholars concerning his choice of methods and sources. That, however, did not discourage Heckscher from extending his study of the Swedish standard of living up to the end of the Napoleonic wars. His entire series is contained in An Economic History of Sweden (Heckscher 1954), posthumously published in English in a translation by the late Go¨ran Ohlin, nephew of Bertil and a fine economist and economic historian in his own right. Mats Morell has carefully examined Heckscher’s work concerning Swedish food consumption, and in chapter 18 he presents a critical evaluation of the methods Heckscher used. Heckscher opted for a physiological approach, weighting various foodstuffs by their calorie content, over the more economics-based real-income approach that relies on incomes and prices. According to Morell, his preference resulted from his extensive research on food and calorie requirements during the blockade-induced food shortage Sweden suffered during World War I. Subsequent research has concluded that Heckscher’s handling of his sources, especially his transformation of old measures, was far from accurate. The net effect was that he greatly exaggerated the consumption of calories during the sixteenth century, and thus that the ‘‘barbaric prosperity’’ he claimed to have found never existed. Despite these shortcomings, however, a number of Heckscher’s results are still valid. These include his discovery of the ‘‘vegetabilization’’ of the Swedish diet during the eighteenth century, as well as the observation that diets in preindustrial times consisted largely of vegetable foodstuffs, particularly grains. The Man Eli Heckscher left an indelible imprint on the Swedish academic world, both as a giant in his field and as an academic entrepreneur. He belonged to a pioneering generation of economists, and as the founder of economic history as an academic discipline in Sweden, he remains a
22
Introduction
prominent figure. Heckscher’s extensive scholarly output, as impressive in its breadth as in its depth, makes his research the starting point for later generations of Swedish economic historians even to this day. Numerous stories about the rather eccentric professor have circulated for several generations, and frequent testimony to his teaching skills is found in memoirs and biographies. Heckscher himself, however, wrote no memoir and a comprehensive biography still remains to be written. The reminiscences that comprise the last section of the book are intended to present the human side of Eli Heckscher. In descending order of age, the grandchildren Eva, Einar, Sten, and Ivar relate their personal childhood memories of their grandfather in chapter 19. They confirm that he was a true workaholic, a continuously working scholar with set routines and a distinctly ascetic personality. Other than at family meals, when visiting their grandparents in Baldersgatan, they always met their grandfather in his study where Eli presided at his desk and the children were invited to sit briefly on his lap. Ivar relates that even those visits were dominated by ritual: first a survey of the items on the desk and then the offer of a piece of candy—the one at the top. No digging through the box in search of a favorite was permitted. On these occasions Eli could be rather distracted and have difficulty totally abandoning his work. A serious accident was narrowly averted when young Einar, on a visit to his grandfather’s study during World War II, was allowed to play with a loaded revolver so that Eli could continue his labors undisturbed. Of course, a shot went off. Eli’s comment, ‘‘Well Einar, I think you had better go and play with grandma for a while,’’ displays a degree of self-control that says a good deal about his character. Indeed, so does the fact that he kept a loaded revolver on his desk, a remarkably un-Swedish practice. The older grandchildren, Eva and Einar, were teenagers, sixteen and fourteen years old, when Eli died in 1952. Thus they have a somewhat more mature perspective on their grandfather than that of a child. Their images are different, but it is possible to merge them into a single portrait. Einar did not live up to Eli’s strict requirements for hard work and success in school. As a result, their relationship became cold and impersonal. Eva, on the other hand, became very close to her grandfather, even though they seldom talked about personal matters. Rather, their frequent conversations concerned issues. In these discussions, regardless of his true convictions, Eli frequently took positions at odds with Eva’s view simply in order to encourage her to think. He
Introduction
23
remained a demanding teacher and an intellectual guide his entire life, even towards his grandchildren. Interestingly enough, the adult Heckscher’s reluctance to speak of himself and his preferences also can be detected in the school essay his grandchild Ivar discovered stashed away. The topic assigned by his teacher encouraged the pupils to reflect personally on their future choice of profession. Sixteen-year-old Eli, however, responded with a totally impersonal discussion of the general factors that influence a person’s choice of profession, with no hint whatsoever as to where his own ambitions lay. His skill as a storyteller is attested to by Sten, the second youngest of the grandchildren: ‘‘We small children sat at his feet and listened to fascinating stories based on his scholarly research.’’ The grandfather he knew, of course, was elderly. Sten Heckscher’s description of his grandfather as a singular mixture of warmth and intimacy, on the one hand, and distance and analytical restraint, on the other, reveals personality traits that also had been noted by contemporaries earlier in Eli’s life. Together with David Davidson, Knut Wicksell, and Gustav Cassel, Eli F. Heckscher has come to be regarded as a member of the first generation of modern Swedish academic economists by virtue of the Heckscher-Ohlin theory. At the time, however, Heckscher was best known as a participant in economic policy debates. In a seemingly endless series of newspaper articles, most of them published in the leading liberal morning paper Dagens Nyheter, he analyzed and commented on current problems. Moreover, he was heard repeatedly on the radio, the revolutionary new medium of the period, as a lecturer and debater. Nonetheless, it is principally as an economic historian that Heckscher became known to posterity. This is clearly demonstrated in chapter 20 where Bo Sandelin examines how frequently Eli Heckscher still is cited in scholarly journals. Between 1986 and 2002, Heckscher’s principal publications in economic history, Mercantilism, Sveriges ekonomiska historia, and An Economic History of Sweden, accounted for more than two-thirds of all his citations, while less than one-fifth referred to his publications in economics. Of the latter, almost all were citations of his the seminal 1919 article. As Sandelin stresses, Eli Heckscher’s citation frequency as measured by the ISI databases, the Social Science Citation Index, and the Arts and Humanities Citation Index, seriously understates his continuing
24
Introduction
scholarly importance. These sources are biased against Heckscher for a number of reasons, chiefly because they are based largely on English language journals and thus fail to provide a worldwide result. In addition, the foreign trade article presents special problems. From the database available it is impossible to tell if these citations refer to the Heckscher article or Bertil Ohlin’s thesis, both of which are included in Heckscher-Ohlin Trade Theory (Heckscher and Ohlin 1991), or to Harry Flam and June Flanders’s introductory chapter in the same volume. Even allowing for these shortcomings, however, the results are not without interest. Heckscher’s most cited work is Mercantilism, which during the 1930s was published in Swedish, English, German, Spanish, and Italian. Two subsequent English editions appeared in 1935 and 1994. Compared to other contemporaneous Swedish economists of the first or second generation, such as Knut Wicksell, Gustav Cassel, Bertil Ohlin, Gunnar Myrdal, and Erik Lindahl, Heckscher places third, ahead of Ohlin, Cassel, and Lindahl, but behind Myrdal and Wicksell. In chapter 21, Benny Carlson deals with Eli Heckscher’s political conversion, which occurred during the 1910s. Heckscher’s image is that of one of the most liberal Swedish economists who never deviated an inch from his free trade position. As a young man before World War I, however, he had appeared on the national scene as a conservative. Indeed, he verged on being a state socialist, advocating government intervention in the economy. At the time he was strongly influenced by his teacher and mentor Harald Hja¨rne, who was a leading conservative thinker. Carlson bases his analysis of Hecksher’s conversion from social conservatism to economic liberalism on a number of sources. These include the testimony of colleagues and other associates, Heckscher’s own published writings, and even his personal correspondence with his mother. Heckscher himself believed that his study of economics, together with the unsatisfactory experience with wartime socialism, was decisive for his altered political views. This is also the conclusion that scholars and other observers have tended to reach. After reviewing Heckscher’s writings on the state, however, Carlson argues that Heckscher actually changed his position sometime during 1910 or 1911, a timing that may well be associated with the franchise reform of 1909 and the shift from a right wing to a leftist government in 1911. Heckscher’s views on Jewish assimilation and Zionism are examined by Harry Flam in chapter 22. His well-known credo that he considered
Introduction
25
himself ‘‘first and foremost as a citizen of Western society . . . only second as a Swede, and third as a Jew’’ did not prevent him from writing about and discussing the so-called Jewish Question. He was a secondgeneration Swede, an assimilated Jew, raised in an economically comfortable and socially well-established family, and married to a Swedish gentile. Like most of his generation of Swedish Jews, Eli Heckscher was totally integrated into Swedish society and did not consciously identify with any Jewish culture or nationality. Heckscher believed that the promotion of Western civilization and its scholarly expression in an unfettered search for truth was the highest of all callings. Jews, just like everyone else, should contribute to that quest rather than to the nationalism that was rampant throughout Europe after World War I. Thus, he argued, the establishment of a Jewish state in Palestine would be disastrous. It would increase anti-Semitism in the nations of the diaspora and create a bloodbath in Palestine. Instead of creating a Zionist state, Jews, in his opinion, should support Western civilization by continuing the process of assimilation. Heckscher was a rational intellectual and moralist, a man of principle, whose worldview was consistently antinationalistic, internationalist, and multicultural. Eli F. Heckscher is one of the greatest Swedish names in the social sciences. For his entire academic career, spanning over fifty years, Heckscher straddled two disciplines, economics and economic history, and during these five decades he was constantly striving to find a way to produce a successful crossbreeding between the two. He did not always succeed, and sometimes he voluntarily abstained from making the attempt. Nevertheless, the plea for incorporating economic theory in economic history, and its mirror image, the plea to make use of history in economic theorizing, were always present in his thinking. The chapters in this book are intended to reflect this preoccupation. By bringing economists and economic historians together we have endeavored to work in the spirit of Eli Heckscher, echoing and resounding his pleas to join economic history in economic analysis, and vice versa. Notes 1. This section draws on Montgomery (1953), Henriksson (1979, 1990, and 1991), and Henriksson and Lundahl (2003a).
26
Introduction
2. A licentiat degree was a preliminary step on the road to the Ph.D. Usually, obtaining it would take three or four years after the fil. kand. (B.A.). It involved the writing of a thesis which had to be defended at an internal departmental seminar. Only a handful of carbon-copied, or even handwritten, versions of the works were made, and basically the degree was an affair between the professor and the student. Hardly any coursework was involved, but the main activity other than the thesis took place in the so-called higher seminar (research seminar). 3. Heckscher’s most important methodological papers are collected in Henriksson and Lundahl 2003b. 4. The other three dealt with exchange rate determination under a paper standard (Heckscher 1916), with a criticism of Gustav Cassel’s purchasing power parity doctrine (continued in Heckscher 1930), Wicksell’s cumulative process (Heckscher 1921) and fixed capital and free goods (Heckscher 1924). Heckscher’s (1918) book on Swedish production problems also contain interesting chapters on overcapitalization and cooperative firms, and in the chapter on tariffs and trusts (originally Heckscher 1913), he anticipates Wilfred Salter’s (1960) vintage approach to capital formation and technological change. 5. Ruben Rausing, unpublished and undated manuscript (ca. 1966), archives of the Stockholm School of Economics. 6. Heckscher (1954) gives an idea about how Heckscher conceived what the final product would have looked like.
References Bergendal, K., et al., eds. 1930. Sweden, Norway, Denmark and Iceland in the World War, Part III. New Haven and London: Carnegie Endowment for International Peace. Eli F. Heckschers bibliografi 1897–1949. Stockholm: Ekonomisk-historiska institutet. Findlay, R., and M. Lundahl. 2002. ‘‘Toward a Factor Proportions Approach to Economic History: Population , Precious Metals, and Prices from the Black Death to the Price Revolution.’’ In R. Findlay, L. Jonung, and M. Lundahl, eds., Bertil Ohlin: A Centennial Celebration (1899–1999), 495–528. Cambridge, MA: MIT Press. Gale, D., and H. Nikaido. 1965. ‘‘The Jacobian Matrix and the Global Univalence of Mappings.’’ Matematische Annalen, 159:81–93. Heckscher, E. F. 1904. ‘‘Ekonomisk historia: na˚gra antydningar.’’ Historisk Tidskrift 24:167–198. ———. 1907. Till belysning af ja¨rnva¨garnas betydelse fo¨r Sveriges ekonomiska utveckling. Stockholm: Centraltryckeriet. ———. 1908. ‘‘Produktplakatet och dess fo¨rutsa¨ttningar. Bidrag till merkantilismens historia i Sverige.’’ In Historiska studier: tilla¨gnade Harald Hja¨rne pa˚ hans sextioa˚rsdag den 2 maj 1908, 693–784. Uppsala: Almqvist & Wiksell. ———. 1913. ‘‘Tullpolitik och trustva¨sen. Med anledning af fo¨rslaget om sockertullens sa¨nkande.’’ Ekonomisk Tidskrift 15:1–29. ———. 1915. Va¨rldskrigets ekonomi. Stockholm: P. A. Norstedt.
Introduction
27
———. 1916. ‘‘Va¨xelkursens grundval vid pappermyntfot.’’ Ekonomisk Tidskrift 18:309– 312. ———. 1918. Svenska produktionsproblem. Stockholm: Albert Bonniers. ———. 1919. ‘‘Utrikeshandelns verkan pa˚ inkomstfo¨rdelningen. Na˚gra teoretiska grundlinjer.’’ Ekonomisk Tidskrift 21, part II: 1–32. ———. 1921. ‘‘Verkan av fo¨r la˚g ra¨ntefot.’’ Ekonomisk Tidskrift 23:49–56. ———. 1922. The Continental System: An Economic Interpretation. Oxford: Clarendon Press. ———. 1924. ‘‘Intermittent fria nyttigheter.’’ Ekonomisk Tidskrift 26:41–54. ———. 1931. Industrialismen. Den ekonomiska utvecklingen 1750–1914. Stockholm: P. A. Norstedt. ———. 1930. ‘‘Monetary History from 1914–1924 in Its Relations to Foreign Trade and Shipping.’’ In K. Bergendal et al., eds., Sweden, Norway, Denmark and Iceland in the World War, Part III, 127–268. New Haven and London: Carnegie Endowment for International Peace. ———. 1935a. Mercantilism. London: George Allen & Unwin. ———. 1935b. Sveriges ekonomiska historia fra˚n Gustav Vasa. Fo¨rsta delen. Fo¨re Frihetstiden. Medeltidshusha˚llningens organisering 1520–1600. Stockholm: Albert Bonniers. ———. 1936. Sveriges ekonomiska historia fra˚n Gustav Vasa. Fo¨rsta delen. Fo¨re Frihetstiden. Andra boken. Husha˚llningen under internationell pa˚verkan 1600–1720. Stockholm: Albert Bonniers. ———. 1949a. Sveriges ekonomiska historia fra˚n Gutav Vasa. Andra delen. Det moderna Sveriges grundla¨ggning. Fo¨rsta halvbandet. Stockholm: Albert Bonniers. ———. 1949b. Sveriges ekonomiska historia fra˚n Gustav Vasa. Andra delen. Det moderna Sveriges grundla¨ggning. Andra halvbandet. Stockholm: Albert Bonniers. ———. 1954. An Economic History of Sweden. Cambridge, MA: Harvard University Press. ———. 1991. ‘‘The Effect of Foreign Trade on the Distribution of Income.’’ In E. F. Heckscher and B. Ohlin, Heckscher-Ohlin Trade Theory, 39–69. Cambridge, MA: MIT Press. Heckscher, E. F., and B. Ohlin. 1991. Heckscher-Ohlin Trade Theory. Cambridge, MA: MIT Press. Henriksson, R. G. H. 1979. ‘‘Eli Heckscher och svensk nationalekonomi.’’ Ekonomisk Debatt 7:510–520. ———. 1990. ‘‘Eli Heckscher.’’ In C. Jonung and A.-C. Sta˚hlberg, Ekonomportra¨tt: svenska ekonomer under 300 a˚r, 165–186. Stockholm: SNS Fo¨rlag. ———. 1991. ‘‘Eli F. Heckscher: The Economic Historian as Economist.’’ In B. Sandelin, ed., The History of Swedish Economic Thought, 141–167. London: Routledge. Henriksson, R. G. H., and M. Lundahl. 2003a. ‘‘Eli Heckscher, ekonomisk teori och ekonomisk historia.’’ In R. G. H. Henriksson and M. Lundahl, eds., Janusansiktet Eli Heckscher: nationalekonom och ekonomisk historiker, 17–49. Stockholm: Timbro.
28
Introduction
———, eds., 2003b. Janusansiktet Eli Heckscher: Nationalekonom och ekonomisk historiker. Stockholm: Timbro. Montgomery, A. 1953. ‘‘Eli Heckscher som vetenskapsman.’’ Ekonomisk Tidskrift 55:149– 185. Olsson, C.-A. 1992. ‘‘Eli Heckscher and the Problem of Synthesis. A Methodological Note.’’ Scandinavian Economic History Review 40:29–52. Salter, W. E. G. 1960. Productivity and Technical Change. Cambridge: Cambridge University Press.
Part I
Theory, Methodology, and History
2
On the Continuing Need for Both the Critical Evaluation of Sources and the Utilization of Economic Theory: Recent Methodological Developments in Economic History Ha˚kan Lindgren
Nearly one hundred years have come and gone since the then twentyfour-year-old Eli F. Heckscher published a paper in the Swedish journal Historisk Tidskrift that introduced and justified a new social science discipline—economic history. Even today, Heckscher’s manifesto seems remarkably fresh, and it has definitely retained its relevance. He positions himself midway between history and economics while skillfully managing to distance himself from the extremes of both fields. The German historical school is criticized both for ‘‘its extreme historical’’ treatment of economic events and developments, and for its ‘‘nomothetic’’ historical writing. The former all too often loses itself in a sea of detail while the latter, within the context of various stage theories of development, results in history being squeezed ‘‘into all too easily manipulated formulas.’’1 According Heckscher, deductive economics suffers from an ‘‘absolute blindness towards the concept of development’’ and a total lack of ‘‘historical perspective.’’ Economic regularities and theories are not universal natural laws, but rather historical categories. They are determined, and will change with, the development of society and the ‘‘living conditions of classes of people.’’ When previous theories and earlier economic policies are treated, it is usually to illustrate the mistakes of the past. Economists ‘‘elevate . . . the principles of free competition to ethical doctrines and take every opportunity to express their purely moral indignation at the crimes that have been committed since the beginning of time against the orthodox teachings.’’2 Given these methodological shortcomings of history and of economics, Heckscher drew the logical conclusion that the new field should combine the best features of both its parent disciplines. The task of economic history, according to Heckscher, was to examine three
32
Theory, Methodology, and History
aspects of the evolution of economic life: conditions, policy, and perspective. Conditions referred to the development of the economy and the living conditions of the population, and were the heart of the new discipline. The history of economic thought, ‘‘perspectives,’’ also was an essential part of the field. It conveyed knowledge concerning how people, at various points in time, understood the interaction among economic variables. Heckscher used the expression ‘‘how the construction of economic activity evolved.’’ That, in turn, determined what economic policies were pursued. Thus, the young Heckscher was well aware that people’s experiences, values, and ideas concerning the true nature of reality form the basis for how the economy was organized during various historical eras and under different cultural systems.3 This essay poses the question of whether Heckscher’s 1904 manifesto is relevant to the economic historian of today and tomorrow. Methodological developments in the fields of history, economics, and economic history during recent decades are contrasted with Heckscher’s vision. This then forms the background to some reflections on the current relationship between economic history and the two disciplines from which it emerged. The concrete examples and the reality, I assume, are largely Scandinavian. The problems I treat, however, have a global and, to some degree, a timeless character. Micro History and Social Constructivism Within the discipline of history, recent decades have witnessed a distinct shift in the focus of research, a development that also has affected the set of methods used to solve problems. This methodological reorientation is also apparent in a number of other humanistic fields such as archaeology, anthropology, and ethnology. It can be summarized as a declining interest in artifacts as such, accompanied by greater attention to the values and the mental framework that can be inferred from the historical remains. Thus, for example, during the 1960s and 1970s, research in social history had a solid positivistic foundation. It was to a great extent dedicated to the quantification of the characteristics of large social aggregates. Class status was by far the most frequently used such category. The reaction against this quantification and aggregation paradigm that occurred during the 1980s and 1990s received its sustenance from French historical psychology and the Annales school. Historians such as Lucien Febvre, Fernand Braudel, and Emmanuel Le Roy Ladurie led the charge.
On the Continuing Need
33
In Sweden, these new trends were noted during the early 1990s in several newspaper articles contributed by two of the leading professors ¨ sterberg in the field, Rolf Torstendahl of Uppsala University and Eva O of Lund University. Rolf Torstendahl emphasized that anthropologically oriented, historical mentality research was in the process of displacing the previously totally dominant quantitative social history.4 ¨ sterberg, for her part, noted that hermeneutics, anthropological Eva O culture analysis, and the gender theories of women’s studies had gradually shifted the interpretive framework of historical research. Nonetheless, she predicted that both materialism and social theory were about to experience a renaissance.5 Currently, little remains of the old quantitative social history. It has been almost entirely replaced by research on mentalities and perspectives, on attitudes towards time and space, suicide, death and religion, as well as toward other persons. Court records have become a diligently exploited source of material. Correspondence, diaries, and other narrative sources are also utilized, often for textual analysis. These materials yield new knowledge concerning people’s life experiences, such as carnivals, cafe´ socializing, and folklore. Paradoxically enough, the growing interest in cultural analysis and mentality research has resulted in a major restriction in the scope and geographic extent of historical research. The Annales school’s initial demand for a ‘‘histoire totale,’’ in practice, has brought on a renaissance for detailed descriptions, local history, and the so-called micro history. Striving to gain insight into the actions and experiences of individual actors has replaced the goal of producing total history. Ironically enough, the fact that the new sources are analytically demanding and very labor intensive to use, has actually reinforced the micro trend.6 ¨ sterberg brought As a result, now, ten years after Torstendahl and O up the issue, micro history directed at charting people’s perspectives and individual fates is the height of fashion in Swedish history departments. When such analysis has been broadened to include groups and experiences at an ‘‘intermediate level,’’ it has even been endowed with a new name intended to differentiate it from macro and micro history: mezzo history.7 During 1999, Arne Jarrick, professor of history at Stockholm University, initiated a new public debate concerning historical scholarship with an article in one of Sweden’s largest morning newspapers, Svenska Dagbladet.8 Basically, the ensuing discussion concerned what constitutes scholarly research and writing in history. Is the search for truth,
34
Theory, Methodology, and History
with a capital T, the essential mission? Indeed, is there a truth to be found, or are our interpretations of various events only cultural constructions that are determined, or in any case influenced, by our own perspectives and values? The answers offered to these questions depended greatly on the writer’s epistemological and theoretical address. To the outsider, the role of the search for truth in legitimizing history as a scholarly discipline might seem trivial. On further reflection, however, it is a highly relevant, and perpetual, problem for all scholarship that seeks to investigate so-called reality. At one extreme, there are scholars who emphasize the primacy of the search for truth and the critical evaluation of sources in order to separate fact from fiction. Even though most modern historians distance themselves from the ‘‘primitive positivism’’ of Comte, Carnap, and Neurath, it is not totally incorrect to refer to this group of scholars as being inspired by positivism. The opposite extreme is represented by modern idealism. In technical jargon, it has been labeled ‘‘social constructivism,’’ ‘‘postmodern relativism,’’ or even ‘‘postmodern constructivism.’’ The extreme relativists maintain that every statement concerning reality is a social construct and that, therefore, there are as many truths as there are speakers. As Ha˚kan Arvidsson argued in the abovementioned debate, history ‘‘ultimately is a construction and a narrative that bears the imprint of the values, insights and biases of its time.’’ In his contribution, Arvidsson does back away from the extremes by drawing a methodological distinction between the determination of the ‘‘particularities of the past,’’ on one hand, and the merging and interpretation of actual events into a unity, or narrative, on the other hand. In the former case, the critical evaluation of sources is needed, even though in practice it seldom yields simple answers as to what is true or false. In the latter case, however, history is caught up in a spider’s web of relativism. Interpretations of our past will always be followed by reinterpretations, simply because ‘‘history is a communication between the past and the present, between then and now.’’9 Economics: A Social Science Apart The new methodological currents with the cultural disciplines, to various degrees and in various ways, have also influenced the social sciences. This indicates that these changes are an expression of a broad scholarly renewal. They constitute a profound reaction against the ma-
On the Continuing Need
35
terialistic and generalistic direction of research that totally dominated the 1960s and 1970s and that one critic aptly called ‘‘the tyranny of aggregations.’’10 Within the area of economic research, the leadership in this process of renewal undoubtedly has come from business economics. This new discipline expanded rapidly at the existing Swedish universities during the 1980s, and, at the same, was represented at most of the new regional campuses that were established during this time. Measured by the number of students and faculty, business economics quickly grew to become the incomparably largest subject among the social sciences. Its multidisciplinary character gave rise to new questions and new methods for illuminating various aspects of business activity. That part of economic history that deals with business history has especially been inspired by the rapidly expanding research concerning company organization and leadership. Research in business economics is, by definition, inductive and theory-testing, often utilizing case studies of individual enterprises. Its predisposition toward such case studies has helped increase the scholarly legitimacy of studies of ‘‘micro structures,’’ even in other areas and in other disciplines. Nor did research in economics remain unaffected by the shift from macro to micro. During the last few decades, the firewall that previously divided macro from micro theory has been breached. In particular, the tools of micro theory have made major inroads into macro territory. This flowering of microeconomics is clearly related to the shift in paradigms that occurred in the 1970s, when it became apparent that reality no longer agreed with the Keynesian road map. There was no room in Keynesian business cycle theory for the combination of accelerating inflation and increasing unemployment. Instead, this unhappy duo had to be given a new name—stagflation. At the same time, faith in Keynesian growth theory was being undermined by the absence of economic growth in the low income countries, despite the massive capital transfers from the industrial countries that had been ongoing since the 1950s. Simply put, the use of broad aggregates to explain the mechanisms underlying the business cycle and economic growth had failed. Thus, an intensive search for the microeconomic basis of macroeconomic variables was undertaken. Monetarism got wind in its sails, while new approaches such as ‘‘supply side economics’’ and the ‘‘economics of entrepreneurship’’ emerged, as did ‘‘post-Keynesian economics,’’ with ambitions to open up for a more nonorthodox economic theory and ‘‘abductivist’’ views
36
Theory, Methodology, and History
of explanation. At the same time, older, so-called heterodox theories, such as Austrian economics, Schumpeterianism, and institutional economics, reappeared and were adapted to mainstream economics.11 Still, economics has remained relatively impervious to the various schools of postmodern thought. The reason for this resistance is that the discipline, or at least its mainstream, rests on a unifying analytical base, a product of its roots in neoclassical theory. No other social science can rely on such a compact core of axioms and premises as does microeconomics. In combination with a strictly logical, deductive methodology, which is particularly well suited to mathematical and statistical expression, economics has with great success applied its hard core of marginal utility maximization and rational behavior based on self interest to many new and, reformulated old, areas of research. This constant theoretical core has both strengths and weaknesses. The deductive analysis and the strong emphasis on mathematical and statistical models have successfully been applied to new problems, but it has also contributed to isolating economics from the other social sciences. During recent decades, economic analysis has made major incursions into other areas of research, including the economics of education, of the environment, of energy, and of health care. Not surprisingly, other social scientists have not hesitated to criticize what they see as inappropriate intellectual imperialism by economists.12 Ever since deductive methods and mathematical models had their breakthrough in economic analysis during the late 1800s, a debate has raged as to whether these methods actually have screened economics from the very social problem it was intended to solve. Stanley Jevons and Le´on Walras laid the groundwork for the two principal approaches that won wide acceptance and for a long time dominated ‘‘mainstream economics.’’ From Jevons, and later Alfred Marshall, emerged the idea that the principal task of economics is to utilize statistical methods to test theoretically based conclusions with empirical data. For some time, this was referred to as applied economics. Walras, with his strictly deductive and mathematical model of general equilibrium, originated the paradigm that eventually came to dominate research and graduate education: that the function of economic science is to develop its internal theoretical apparatus, ‘‘positive economics,’’ with the help of increasingly refined abstract models and ever more advanced mathematical methods.13 More recently, the ever increasing specialization and ‘‘technification’’ of the subject according to Walrasian principles has led to a new de-
On the Continuing Need
37
bate among economists as to whether or not the discipline puts excessive emphasis on the solution of technical problems at the expense of social relevance. Assar Lindbeck, professor emeritus at Stockholm University, has forcefully warned against economics abandoning the social sciences to become a pure natural science. Lindbeck’s successor in the Stockholm chair in international economics, Lars Calmfors, emphasized several years ago in a discussion paper that technification has increased the distance from empirical reality. In order to give economic scholars perspective and a more relativistic view of their mathematical and statistical models, the history of economic thought and economic history should be required subjects for graduate students, he argued.14 Economics and History At the end of World War II, research and instruction in economic history in Sweden was constrained to one single academic setting: the Institute for Economic History Research that had been founded by Eli Heckscher. Its institutional home was at the Stockholm School of Economics and what was then called Stockholm University College. As part of postwar university reform, however, and thanks to Heckscher’s influence at the ministerial level, the field was given more independence from the departments of history than had been proposed by the prelegislative committee proposals. New readerships in economic history were filled in Lund by Oscar Bjurling (1948), in Gothenburg by Artur Attman (1949), and in Uppsala by Karl-Gustaf Hildebrand (1951). All had been active in the history departments of their respective universities. Heckscher’s own pupil Ernst So¨derlund was appointed to the professorship in Stockholm (1949), and a separate economic history department was created. Of these scholars, only Bjurling was an economist. All the rest had their roots in the field of history. The institutional liberation of the field was completed, first when the readerships were converted into professorships during the 1950s, and second when the humanistic faculties were divided up during the mid-1960s. Since then, training and research in economic history have been located in separate departments of economic history. Over time, they have developed a consensus as to what constitutes good research in the field. In 1966, these departments of economic history were transferred to the faculties of social science, thus, at least in the long run, strengthening their contacts with the other economic disciplines.
38
Theory, Methodology, and History
Courses in time series analysis, growth theory, and the history of economic thought were introduced, but methodologically the field was, and remains, firmly anchored in the inductive empirical tradition.15 A clear indication that the roots of Swedish economic history lay in the cultural field of history is that the discipline remained relatively unaffected by the ‘‘new economic history’’ and the cliometric movement that emerged in the United States during the 1960s and 1970s. The historical economics that flourished among American economic historians at that time was intended to widen the boundaries of neoclassical economics through a conscious application of the logical deductive approach to historical problems. Critics, however, argued that deductive models based on simple assumptions are incompatible with historical research which is intended to explain complicated (multicausal) empirical relationships. The ensuing methodological struggle petered out during the 1980s, with neither side being able to claim victory. The methodological chasm between economic and historical thinking was just too deep.16 As in the other Scandinavian countries, Swedish economic history has been characterized by a relativistic perspective. This has found expression in a marked openness toward new trends, both with regard to theory and to choice of topics. Marxist-inspired research on the transition from one mode of production to another had its day. It was succeeded by work-life research during the 1980s and by women’s history, now called historical gender research, during the 1990s. On a more basic level, however, the methodological unanimity has been great: formal theory, at various levels of abstraction, regardless of origin, is to be tested against a concrete, multifaceted reality. To the law-seeking ‘‘nomotetists’’ this was justified by the argument that a theory continually has to be confronted with reality to prevent if from becoming a shackling dogma. To the empiricists it was argued that detailed descriptions of ‘‘the living past’’ must be imbedded in more general explanations and theories in order to have scientific interest. The epistemological basis of this consensus has been materialistic in the sense that is founded on a belief that a reality, about which it is possible to gain knowledge, exists independent of the individual’s perceptions.17 In recent years, this consensus has been challenged by micro history and postmodernism, the latest fashions in history and the other cultural disciplines. Especially at Stockholm University, where research in social history has had a long tradition, people’s experiences and perceptions has become an area of research within
On the Continuing Need
39
economic history. A number of monographs and theses with a sociological and anthropological bent have recently been written there. Their principal inspiration has been the French cultural sociologist Pierre Bourdieu. One interesting example is the research project entitled ‘‘A History of Human Conceptions.’’ From a social and mentality perspective, it examines questions such as how gender relations, criminals, suicides, and children were perceived in the past.18 These trends within the academic discipline of economic history naturally raise the question of whether or not it has a hard core. What should be considered its central area of concern and what should be viewed as peripheral? Even if we opt for inclusion rather than exclusion, it must still be asked if postmodernism and social constructivism is relevant to the internal development of the discipline. What does a far reaching social constructivist perspective add and what are its dangers? That question is not so easy to answer. As long as I can remember, in our methodology courses we have taught our students that scholarly research in history requires a critical examination of myths and socially useful metaphors. Otherwise our defenses against all sorts of propaganda and biased information are likely to be inadequate. This scholarly attitude is operationalized in various ways: the explicit formulation and analysis of problems, critical and independent thinking, and an understanding of the context. Not least, the requirement of insight and understanding calls for consideration and respect for other points of view and the values of other people. This applies not only to colleagues in one’s own discipline, but also to those who work in adjoining fields, and who pose somewhat different questions. In the historical disciplines, however, it is especially respect for the values and perspectives of previous generations that is essential. Any serious current researcher must, to the best of his or her ability, strive to unravel the cultural code that permeates human relations in all societies. In a recently published book, So¨ren Barlebo Wenneberg, a scholar at the Copenhagen Business School, has analyzed the problems and possibilities of social constructivism in an understandable and easily grasped presentation. He works with four different methodological standpoints, all of which fit within, or are associated with, the concept of social constructivism. The first two of these are in a sense related. The first is simply a critical perspective based on the conviction that social phenomena are the product of human actions, not natural laws. The second consists of various theories of how this social reality is
40
Theory, Methodology, and History
structured and holds together. A third viewpoint concerns epistemology and the nature of knowledge. In its most radical version, it maintains that knowledge and scientific concepts exclusively are determined by social factors (such as ‘‘power’’ and ‘‘interests’’) and subjective factors. The fourth and final position can be regarded as the ultimate consequence of the other three. It rejects realism and supports the idealistic point of view that ‘‘reality’’ (especially social reality) is created by our perception and knowledge of it.19 No historian is likely to have trouble accepting the first two of these methodological positions. Indeed, they are the basis of our belief in path dependence, usually expressed in terms such as saying that some phenomena are ‘‘historically determined’’ and that special theories, such as Hegel’s theory of history, are needed to explain the relationship between continuity and discontinuity in the process of change. In my opinion, however, today’s intellectual fashion of including epistemological and ontological analysis creates a two-pronged danger for historical scholarship. First, if two assertions that contradict each other constitute competing descriptions of reality and can be equally true, depending on the authors’ different perspectives, then logic indicates that all conclusions have scholarly validity. Even those that actually are nothing more than a confirmation of the scholar’s own values could not be excluded.20 Second, there is apparent peril in a one-sided reliance on ‘‘today’s narratives’’ to determine what is of historical importance. Namely, the risk that the norms of our time, or our society, become the basis for value judgments concerning phenomena and events of the past. Writing history backwards also makes it possible to enlist history in the service of political propaganda, something frequently practiced in the past. Without an open mind and an honest intent, the cultural codes of the past will never be deciphered, and without that it will never be possible to understand why people acted as they did in various historical situations. For example, understanding changes over time in the concept of ‘‘childhood,’’ that is, to understand its character as a social construct, is essential for studies of the treatment of children in the past. With such a content, the social constructivist approach has important strengths. That, of course, was the reason that Eli F. Heckscher in his 1904 manifesto so strongly emphasized the key role of the history of thought in economic history. Economic theories are historical categories, and the history of economic thought reveals how, at various
On the Continuing Need
41
points in time, people understood the functioning of the economy. That, in turn, determined their economic behavior. It is, of course, rewarding to study historical actors with hindsight. As Go¨ran B. Nilsson put it, however, it also leads ‘‘to a trivialization and simplification of events.’’ All history, including economic history, should be written forward. It should be based on a holistic perspective, carefully examine its sources, and strive to understand historical actors and their behavior. Knowing in retrospect what actually came to pass, the historian is tempted to ignore the expectations for the future held by his subjects. That, however, would be a mistake. As Nilsson argues, ‘‘the historian should take the anticipations of his subjects seriously, even though he knows better than them.’’21 This prescription constitutes a well-thought-out and articulated alternative to the moralizing history writing that unquestionably has grown in popularity. In fact, of course, the latter is a return to the older history writing that flourished before the breakthrough of ‘‘Rankeism’’ during the nineteenth century. The fundamental rules for scholarly historical research that first were preached in the middle of the nineteenth century by Leopold von Ranke, an influential professor at the University of Berlin for almost half a century (1824–72), are still valid today. They require a critical evaluation of sources and a striving for objectivity, as well as historical understanding. That is to say, the scholar should endeavor to set aside the values of his or her time and attempt to understand and interpret historical events in light of the then existing conditions. This demand to free oneself of one’s own values requires that the scholar not only have a profound knowledge of the historical period being studied— how did people think then?—but also that the subject be approached with considerable humility and respect for the actions of the historical actors. In 1824, Ranke’s manifesto was summarized in an appendix to his first major work, Geschichte der romanischen und germanischen Vo¨lker: no longer was historical research intended to judge the past, but, to the greatest extent possible, to determine what had actually happened, ‘‘wie es eigentlich gewesen.’’22 Economic History as Economic Science Economic history is an inductive, empirical discipline. Even though the purpose of economic historical analysis is to describe reality, to
42
Theory, Methodology, and History
discover repetitive phenomena and to reveal similarities and patterns in economic development, it is not atheoretic. Perhaps its most important task is to test various economic theories, and thereby to contribute to the development of new theoretical constructs concerning the nature of reality. Thus, the field has clear connections to ‘‘applied economics’’ in the spirit of Jevons and Marshall, even though the economic historical method has fundamental differences from that of economics. The latter works with a theoretical-deductive approach to analysis. Thus, economics treats as constants factors that in economic history are considered short-run parameters and long-term variables. As Heckscher emphasized in a 1947 article, the practice in economics of abstracting away portions of economic reality inevitably means that it can not provide the inclusive analysis of events that economic history strives to deliver.23 The extent, if any, to which the abstract deductive approach of economic theory is compatible with pronouncements concerning empirical reality was the subject of a lively debate among nineteenthcentury classical economists. After all, modern economics emerged from seventeenth-century mercantilism, where the point was to determine how economic policy should be formulated so as to increase a nation’s political power. Normative economics, which provides advice and counsel on economic policy, thus has a long tradition. This background is also reflected in the name given to the new discipline; until the end of the nineteenth century it was called political economy. During the nineteenth century, economics slowly freed itself from simply being a provider of policy prescriptions, and became established as a positive discipline, an ‘‘economic science’’ distinct from the ‘‘art of policy.’’ The basic distinction between positive and normative economics is usually credited to Nassau William Senior and his 1836 work An Outline of the Science of Political Economy. In a famous pronouncement, Senior maintained that the conclusions drawn from an economist’s scientific work ‘‘do not authorize him in adding a single syllable of advice.’’ The overriding idea was that reality is so complicated that the conclusions reached through use of deductive economic theory cannot serve as the sole basis for recommendations concerning which actions should be taken in a given situation.24 The idea of divorcing theoretical (that is, scientific) economics from the empirical variant was perhaps argued most forcefully by the Cambridge economist Henry Sidgwick, at the time highly respected but now largely forgotten. In the Introduction to his Principle of Political
On the Continuing Need
43
Economy (1883), he declaimed that ‘‘it is very rarely, if ever, that practical economic questions which are presented to the statesman can be unhesitatingly decided by abstract reasoning from elementary principles. For the right solution of them full and exact knowledge of the facts of the particular case is commonly required; and the difficulty of ascertaining these facts is often such as to prevent the attainment of positive conclusions by any strictly scientific procedure.’’25 During the twentieth century, economics has evolved in the direction of ‘‘pure’’ theory, with an ever-increasing emphasis on deductive models, relying on sophisticated mathematical and statistical methods. This development has increased the methodological chasm separating theoretically oriented economists, not just from economic historians, but also from policy oriented economists. Within one school of thought, the ‘‘Austrian,’’ the discussion concerning whether it is possible to draw empirical conclusions on the basis of deductive theoretical models has been kept alive. A common conviction of all those economists who describe themselves as ‘‘Austrian,’’ or today even ‘‘neoAustrian,’’ is that no discipline that ultimately deals with relations among people can yield exact predictions, such as those generated by physics and other natural sciences. Thus, the apparent exactness of orthodox economic’s mathematical models is misleading. As the basis for economic policy decisions, it can sometimes be catastrophic.26 These difficulties have been clarified by, among others, Friedrich von Hayek. His installation lecture, given in conjunction with his assumption of the chair in economics at the University of Freiburg in 1962, was devoted to an extremely illuminating discussion of the principles underlying the relationship between scientific professionalism and participation in practical decision making. Hayek’s most important justification for drawing a clear boundary between economics and politics was methodological. Abstraction and theorizing are essential for the development of economic science, as ‘‘knowledge of facts as such is not science.’’ By themselves, however, theoretical relationships do not make it possible to predict the results of one or another economic policy action in a given historical situation. The analysis of economic relationships requires not only insight into theoretically deduced relationships, but also knowledge of the actual conditions that in a concrete situation will affect the outcome. This problem affects all science that deals with complex process involving human beings. According to Hayek, however, economists appear to be more prone than other researchers to forget that advanced theoretical insights are insufficient
44
Theory, Methodology, and History
when it comes to predicting the consequences of a particular action in a given situation.27 In my judgment, the methodological tendencies of ‘‘mainstream economics’’ during recent decades have made Hayek’s argument more relevant than ever. As the problem areas subjected to economic analysis have been broadened and given a time dimension, the economist’s tool chest has been further supplemented with mathematical models a la mode. It also seems as if economists actually have become less aware of the limited empirical validity of theoretical reasoning applied to a concrete economic situation. Apparently this has resulted from the increasing study-time-related displacement of graduate courses intended to convey a broad perspective and a realization that real-world phenomena are indeed complex, by ever more technical courses. Much has no doubt been gained by this development, but something has also been lost. None of the above said implies that economic history should dump its economic ballast. On the contrary, as long as the practitioners of both economics and economic history are aware of the basic methodological differences between the fields, and of the limits and the strengths of the deductive and inductive approaches, major synergetic effects may arise. Ultimately, the nature of the research problem addressed is crucial; naturally, not all questions in economic history can be resolved by economic theory. Nonetheless, as long as examining ‘‘how the problem of satisfying needs, broadly defined, has been solved through time’’ remains the central task of economic history, all theorizing concerning the efficient allocation of scarce resources will be of interest. The principal of scarcity, as well as its antithesis, the principal that individuals with given preferences rationally strive to maximize their payoffs, thanks to their simplicity and general applicability, have impressive explanatory power.28 Let me make the benefits of economic theory for research in economic history tangible by giving two examples, one hypothetical and one from a doctoral thesis recently presented here at the Stockholm School of Economics. The price system is crucial for the allocation of resources in a society. Should, let us say, the price of grain greatly increase in a single year, the economic historian with training in economics immediately will know where to look. He or she will analyze the phenomena in light of the general principles of supply and demand, consider the functioning of the market, and give thought to the substitutability among various grains and of grains with other food-
On the Continuing Need
45
stuffs, both in production and in consumption. Thus, abstract economic theory, although not the entire explanation, becomes the gateway to an explanation. For a complete and exhaustive explanation, it is at least as important to know if the price rise resulted from an increase in demand for grain as a result of a domestic potato crop failure or from a general crop failure in North America. In his thesis, entitled ‘‘Empirical Studies in Money, Credit and Bank¨ gren studies the nineteenth century Swedish financial ing,’’ Anders O system, which was based on private note issuing banks. One of the questions he poses is how the rapid expansion of the country’s money supply could have been compatible with the fixed change rate that the Swedish central bank maintained between 1834 and 1913. In answering this question, the thesis exemplifies not only how good empirical research requires a firm grasp of both economic theory and of econometric methods, but also the importance of historical knowledge ¨ gren’s gateway to an concerning how things actually functioned. O analysis of the functioning of the specie standard in an international perspective is various economic theories of monetary and exchange rate systems. He then utilizes econometric techniques to explain how central bank reserves, the money supply, and prices interacted. The results of his causality tests indicate that reserves affected the money supply, which in turn affected the price level, all in accord with the classical specie flow mechanism. At the same time, both Swedish prices and central bank reserves moved in general accord with those of the country’s principal trading partners. Supplementing the analysis with empirical information leads to the conclusion that the specie standard system was more flexible than predicted by theory. This result was a consequence of the Swedish central bank’s policy of accumulating foreign assets, which could be drawn upon when domestic reserves shrunk. The international connection was of decisive importance for the exchange rate regime. Prices followed the monetary expansion, but so did economic growth, and they were all international phenomena during the gold standard era.29 To Be a Historian While Remaining an Economist In this essay, I have shown how the methodological developments of the two fields that have traditionally constituted economic history’s closest neighbors, history and economics, have developed in diametrically opposite directions. The growing interest among historians in
46
Theory, Methodology, and History
mentality and cultural analysis has given social constructivism and micro history wind in their sails. By contrast, economists have stuck with their hard core and have further developed the strictly logical deductive approach with the help of increasingly advanced mathematical and statistical techniques. The instinctive conclusion ought to be that economic history has arrived ‘‘at the end of the road’’ and now has to choose its scholarly partner. Is it now even possible to maintain, along with Eli F. Heckscher, that an economic historian both must ‘‘be familiar with the abstract method and habit of reasoning on the basis of simplified assumptions of theoretical economics’’ and also ‘‘have the historians ability to synthesize and his sensibility to the multifaceted nature of reality’’?30 I would argue that it is perfectly possible. Social constructivism, in the sense that there are ‘‘cultural codes’’ that must be deciphered, is perfectly familiar to historically trained scholars. Thus, for example, in economic history the basic justification for studying the history of economic thought is that people’s conceptions of how reality is constructed determines the framework for interpretation and action during various epochs. It is only when social constructivism is interpreted in terms of epistemology and ontology that it becomes a danger to historical scholarship (and other disciplines). The problem, above all, is that it opens the door to absolute methodological individualism and subjectivity, as well as a one-sided reliance on ‘‘today’s narratives’’ to determine what is historically meaningful. Materialism and realism must remain the epistemological base of economic history, nor should we revert to the pre–Leopold von Ranke, moralizing style of writing history. Of course the need for insight into abstract economic theory and deductive model building depends on how the field of economic history is defined. As long as the label is ‘‘economic’’ history, however, economic questions must be the heart of the discipline. Other aspects of human activity, no matter how closely related to economics, are secondary. If the broad definition that recurs, in various formulations, throughout Eli F. Heckscher’s writings, that the field concerns ‘‘how human needs have been met throughout time,’’ is maintained, then the principle of scarcity is raised to a general phenomena that all people in all societies at all times have been obliged to face. It is therefore easy to agree with Heckscher when he continues: ‘‘Methodologically speaking, this is the explanation why economic theory is indispensable.’’31
On the Continuing Need
47
A conscious utilization of economic theory in empirical research, however, does not excuse the economic historian from the customary critical evaluation of sources. It is thus essential to be fully aware of the assumptions underlying economic theory and of the limits on the conclusions drawn from deductive reasoning when they are confronted with a complicated and multicausal reality. Such an empirical confrontation, however, can also serve to enrich theory. Thus, economic history can make important contributions to economic theory and, indeed, can be developed in a theory-generating direction. It is important to emphasize that my pleading in favor of a deliberate rapprochement with economic theory does not mean that I in any sense find it desirable for economic history to evolve into a branch of economics. Economic history is by no means historical economics. The boundary is crystal clear. Economic history is an empirical, inductive discipline, deeply rooted in historical method and the critical evaluation of sources. What I do hope, however, is that the trend toward estrangement from economic theory and growing distaste for econometric modeling, which recently can be observed within the field in Sweden, will be broken. Making the application of economic theory explicit and more energetic, however, requires a considerable increase in the qualifications of our economic historians. The next generation of Swedish economic historians must acquire a considerably better knowledge of modern economic theory and of the history of economic thought than is possessed by the current generation. Graduate training in economic history is still dominated by the humanistic educational tradition, with long reading lists and insignificant training in economic theory and methodology. It is the duty of the currently active generation to see to it that graduate education in the field be at least partly reorganized. Multidisciplinary methodology and theory courses, as well as training in quantitative methods, should be given a substantially larger role. If this can be accomplished within the framework of national graduate course cooperation, it simultaneously would satisfy both myself and the evaluation section of the Swedish National Agency for Higher Education. Notes The title of this essay is an echo of that used by Go¨ran B. Nilsson thirty years ago in Historisk Tidskrift (1973: 173–211). That paper opened my eyes to one of the central problems in historical research, and I am pleased to be able to associate my essay with his work.
48
Theory, Methodology, and History
1. Hecksher (1904: 188, 197). 2. Hecksher (1904: 169, 173, 185). In current usage, one would say that economic theories are ‘‘social constructs’’ or ‘‘socially and historically embedded.’’ 3. Hecksher (1904: 185–186). 4. Torstendahl (1991: 1). ¨ sterberg (1992: 24). 5. O 6. Ho¨gskoleverket (2003: 12); Tosh (2000: 110–118, 240–242). 7. Ode´n (1998: 11). 8. Jarrick (1999: 16–17); Nordqvist and Wiklund (1999: 12); Magnusson (1999: 18); Arvidsson (1999: 14). 9. Arvidsson (1999: 14). 10. Dahme´n (1979: 256). 11. Dahme´n (1979: 258–262); Lindgren (2003: 14); Sellstedt (2002: 131–133). 12. Ho¨gskoleverket (2002: 17–29); Bergstro¨m (1996: 77–78). 13. Landreth and Colander (1994: 231, 236, 266–281). 14. Calmfors (1996: 239–240); Lindbeck (2001: 31). 15. Ho¨gskoleverket (2003: 30–32); Hettne (1980: 151–155). 16. Ho¨gskoleverket (2003: 30–32); Hettne (1980: 146). ¨ sterberg is presumably using the term ‘‘materialism’’ in this sense. 17. Eva O 18. Hedenborg (1997: 22–23). For theses inspired by Bourdieu, see Gustavsson (2002, especially pages 10–27), and Eriksson (2004: 9–27, 245–249). 19. Wenneberg (2001: 62, 69–70, 80–81, 92–93). 20. This, of course, does not constitute a denial of the well-known fact that our view of reality affects the construction of scholarly concepts and theories. The complexity and multifaceted nature of reality makes it highly likely that two scholars, working independently, will produce two different descriptions of a given aspect of reality, both of which are ‘‘reasonably true.’’ What is more, these descriptions are more likely to be complementary than competitive in nature. Thus, the constant rewriting of history is not a result of arbitrary or relativistic scholarship. Rather, it is the infinitely multifaceted and multidimensional state of reality that enables the scholarly community to constantly produce new insights into the past as society evolves and the focus of interest shifts. 21. Nilsson (1989: 3); Nilsson (1990: 67). 22. Torstendahl and Nybom (1988: 25–36). 23. Heckscher (1947: 3–4). 24. Schumpeter (1954: 536–541). It should be noted that the classic distinction between normative and positive economics has nothing whatsoever to do with the role of values in scholarly work. For historians and historically trained economists, the importance of
On the Continuing Need
49
the attitudes and the concept of reality prevalent during various epochs is so apparent that it would never occur to them that positive economics is value-free. 25. Reproduced in Schumpeter (1954: 805 n. 11). 26. Hakelius (1995: 31–37, 62–76). 27. Hayek (1967: 259–265). 28. The quotation is from Hecksher (1922: 15). ¨ gren (2003: 177–212). 29. O 30. Hecksher (1922: 36). 31. Hecksher (1936: 17).
References Arvidsson, H. 1999. ‘‘Historiker ma˚ste konfrontera da med nu.’’ Svenska Dagbladet, August 19. Bergstro¨m, V. 1996. ‘‘Ar ekonomerna samha¨llsvetare?’’ In L. Jonung, ed., Ekonomerna i debatten—go¨r de na˚gon nytta? Stockholm: IVA and Ekerlids. Calmfors, L. 1996. ‘‘Nationalekonomernas roll under det senaste decenniet—vilka ar la¨rdomarna?’’ In L. Jonung, ed., Ekonomerna i debatten—gor de nagon nytta? Stockholm: IVA and Ekerlids. Dahme´n, E. 1979. ‘‘Kan den fo¨retagshistoriska forskningen bidra till den ekonomiska teoriens utveckling?’’ Historisk Tidskrift 99, no. 3: 255. Eriksson, P. 2004. Stadshypoteks plats och bona inom det svenska kreditva¨sendet 1909–1970— en socialhistorisk studie. Stockholm: Almqvist & Wiksell. Gustavsson, M. 2002. ‘‘Makt och konstsmak. Sociala och politiska motsa¨ttningar pa˚ den svenska konstmarknaden 1920–1960.’’ Ph.D. diss., University of Stockholm. Hayek, F. A. 1967. ‘‘The Economy, Science, and Politics.’’ In Studies in Philosophy, Politics and Economics. London: Routledge & Kegan Paul. Heckscher, E. F. 1904. ‘‘Ekonomisk historia. Na˚gra antydningar.’’ Historisk Tidskrift 23, no. 4: 13–198. ———. 1922. ‘‘Ekonomi och historia.’’ In Ekonomi och Historia, 11–50. Stockholm: Albert Bonniers. ———. 1936. ‘‘Den ekonomiska historiens aspekter.’’ In Ekonomisk-historiska Studier, 9–69. Stockholm: Albert Bonniers. ———. 1947. ‘‘Ekonomisk historia och dess gra¨nsvetenskaper.’’ Historisk Tidskrift 67, no. 1: 1–17. Hedenborg, S. 1997. Det ga˚tfulla folket. Barns villkor och uppfattningar av barnet i 1700-talets. Stockholm: Almqvist and Wiksell. Hettne, B. 1980. ‘‘Ekonomisk historia i Sverige under femtio a˚r. Institutionell utveckling och forskningsinriktning.’’ Historisk Tidskrift 100, no. 1–2: 140–175.
50
Theory, Methodology, and History
Ho¨gskoleverket (National Agency for Higher Education). 2002. Utva¨rdering av utbildning i nationalekonomi vid svenska universitet och ho¨gskolor. Ho¨gskoleverkets rapportserie 2002:9 R. ———. 2003a. Utva¨rdering av a¨mnet ekonomisk historia vid svenska universitet. Ho¨gskoleverkets rapportserie 2003:11 R. ———. 2003b. Utva¨rdering av grundutbildning och forskarutbildning i historia vid svenska universitet och ho¨gskolor. Ho¨gskoleverkets rapportserie 2003:12 R. Jarrick, A. 1999. ‘‘Vetenskapen hotar fo¨rfalla till kollektiv monolog.’’ Svenska Dagbladet, June 20. Jonung, L. 1996. ‘‘Inledning.’’ In L. Jonung, ed., Ekonomerna i debatten—go¨r de na˚gon nytta? Stockholm: IVA and Ekerlids. Landreth, H., and D. C. Colander. 1994. History of Economic Thought. Boston: Houghton Mifflin. Lindbeck, A. 2001. ‘‘Economics in Europe.’’ CESifo Forum (Spring 2001): 31–32. Lindgren, H. 2003. ‘‘Scandinavian Business History at the End of the 1990s: Its Prior Development, Present Situation, and Future.’’ In F. Amatori and G. Jones, eds., Business History Around the World, 146–169. Oxford: Oxford University Press. Magnusson, L. 1999. ‘‘Allsko¨ns vidskepelse framkallar rysningar.’’ Svenska Dagbladet, August 11. Nilsson, G. B. 1989. ‘‘Historia som humaniora.’’ Historisk Tidskrift 109, no 1: 1–15. ———. 1990. Den lycklige humanisten. Tio offensiva essa¨er. Stockholm: Carissons. Nordqvist, S., and M. Wiklund. 1999. ‘‘Historikerna ma˚ste bidra till nuets bera¨ttelser.’’ Svenska Dagbladet, July 16. Oden, B. 1998. Leda vid livet: Fyra mikrohistoriska essa¨er om sja¨lvmordets historia. Lund: Historiska Media. ¨ gren, A. 2003. ‘‘Empirical Studies in Money, Credit and Banking. The Swedish Credit O Market in Transition under the Silver and Gold Standards, 1834–1913.’’ Ph.D. diss., Stockholm School of Economics. Osterberg, E. 1992. ‘‘Historieforskare, lyft na¨san en smula!’’ Svenska Dagbladet, November 11. Schumpeter, J. A. 1954. History of Economic Analysis. London: Allen & Unwin. Sellstedt, B. 2002. ‘‘Metodologi fo¨r fo¨retagsekonomer. Ett fo¨rso¨k till positionsbesta¨mning.’’ Stockholm: SSE/EFI Working Paper Series in Business Administration. Available online: http://swoba.hhs.se/hastba/abs/hastba2002_007.htm. Torstendahl, R. 1991. ‘‘Historieforskning: Kultur och makt.’’ Svenska Dagbladet, September 2. Torstendahl, R., and T. Nybom. 1988. Historievetenskap som teori, praktik, ideologi. Stockholm: Fo¨rfattarfo¨rlaget. Tosh, J. 2000. Historisk teori och metod. 3rd ed. Lund: Studentlitteratur.
3
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904* Rolf G. H. Henriksson
Eli Heckscher’s fame as both an economic historian and an economic theorist also entails recognition of him for his methodological position. He was already noted internationally for his writings on the methodological theme in the inter warperiod when he, somewhat in opposition to an early mainstream in the field, made a plea for theory in economic history and followed it up arguing for a more quantitative statistical approach.1 Then in the skirmishes caused by the emergence of the so-called new economic history in the 1960s, he met renewed attention as somewhat of a precursor to that movement (Fogel 1965). The basic motive for the attention in this essay to Heckscher’s methodological views is that comprehending them is central to the task of reassessing also his substantive contributions.2 In 1904, at the outset of his career as an economic historian, Eli Heckscher made a much noted methodological declaration (Heckscher 1904b). Methodological issues concerning the interface of economics and history then remained a central concern for him throughout his life.3 But in a life of research and thinking spanning more than five decades of scholarly pursuits, Heckscher’s methodological views were considerably reformulated. However, in my view the early declaration of 1904 stands as a manifesto, whose core tenets Heckscher never abandoned.4 The present essay seeks to lay bare Heckscher’s conception of the task, scope, and method of the field as stated in the manifesto of 1904. This is done by presenting the manifesto generically as the result of exposures to the fields of history, political science, and economics, which were Heckscher’s three main subjects during his study years at Uppsala University (1897–1904). By especially scrutinizing more closely the background of the manifesto it is possible to interpret its content and message more deeply than has so far been done.
52
Theory, Methodology, and History
The content and argument of the essay is as follows: a first lengthy section, reviewing Heckscher’s various lines of study, seeks to clarify his starting position before the 1904 declaration as mainly that of a general historian. However, it points out that already in this phase Heckscher had received at least elementary research experiences in social science that prepared for his later breakaway from the established track of historians. He had familiarized himself with the statistical approach in political science and the conceptual approach in economics. The next section offers an appraisal of Heckscher’s work with the licentiate thesis (Heckscher 1903; 1908a). Here Heckscher, in his first major research project as a historian, is seen to have arrived at an intermediary position in his transition from history to economic history. As the thesis is but a first synoptic anticipation of his later work on mercantilism and merely an embryo of his Economic History of Sweden, he had not yet fully conceived of how to employ either the statistical or the conceptual approaches that his exposure to the two social sciences might have suggested. The subsequent section deals directly with the manifesto of 1904. Here Heckscher took the decisive step in assuming an identity as an economic historian. The account expounds on and offers an interpretation of the content of the manifesto as regards the roles of economics and other social sciences in defining the task, scope, and method of economic history. A brief conclusion sums up the survey and the argument. Apprentice in Uppsala When Heckscher, at age eighteen, entered Uppsala University, in early 1897 he first spent some time shopping around and orienting himself in academic life before settling for his chosen line of study. After a while, he reported to his parents that he had never dreamed that he would earn his first income as an actor in a play. This public activity of Heckscher illuminates a passionate side of his personality that may easily be forgotten in an intellectual portrayal that will in the main underscore the rational and systematic way in which this very focused man, dedicated to high moral principles, sought to go about his duties and solve the problems he encountered. Heckscher had graduated from the so-called Latin line at Nya Elementar, the high school in Stockholm where he had prepared himself
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
53
for the university. This line aimed quite exclusively for humanistic and law subjects rather than ‘‘sciences.’’ Facing his academic options thus confined, Heckscher seems at first to have considered mainly linguistic subjects in his study program.5 However, Heckscher’s basic inclinations were early bent towards historical studies. He appears to have made a predisposing acquaintance with that field already during his next to last year at Nya Elementar. In the summer of 1895 he attended courses at Uppsala University, where he gained his first impressions of a number of leading university professors, including the historian Harald Hja¨rne, which as a sanguine father figure was later to become his main mentor.6 Although, Heckscher appears to have spent considerable time before finally committing himself to a study program he had already in his first year at Uppsala enrolled in historical studies and did seminar work assigned by Hja¨rne. However, these studies under Hja¨rne were broken off in the spring of 1898 by a sojourn at Gothenburg University.7 Here he had the opportunity to pursue studies under the Ludvig Stavenov, who in 1895 had been appointed to the chair of history and politics set up at that new university.8 Stavenow had less of the flashing brilliance that was the prime mark of Hja¨rne’s excellence, but was nevertheless a historian of wide perspectives and intuition who no doubt influenced Heckscher’s later thesis work for the fil. lic. degree. Thus history became Heckscher’s primary choice and hence to be the subject where he wrote a thesis. On returning to Uppsala in the fall of 1898 he was more tightly drawn into the circle of historians gathering around Hja¨rne. This socializing with the Hja¨rne group appears to have been the decisive influence not only for his studies in history but for his other studies as well.9 Of special importance was the friendship Heckscher made with a more advanced history student, who like himself was a graduate of Nya Elementar, the charismatic Gunnar Hazelius.10 Hazelius was a leading personality in Uppsala student life in the 1890s and exerted an intellectually as well as morally kindling influence on student minds in a wide network. Hazelius seems to have played an important role in guiding Heckscher on his initial study program. Apparently Hazelius was the major influence in making Heckscher the first in the circle of history students to formally sign up for economics in a fil. lic. degree as an option available after a university reform act of 1891. On recommendation from Hazelius, who had no doubt read Wicksell’s favorable review, Gide’s Nationalekonomins Grunddrag (1899) seems to have been the primer for Heckscher’s
54
Theory, Methodology, and History
studies in economics.11 That study was a newly conducted translation into Swedish of a French classic that offered, as far as a text book goes, a solid up-to-date overview of the state of economic science, covering also the recent advent of neoclassical theory. Such a textbook introduction was no doubt important to Heckscher, inasmuch as economics was at Swedish universities only taught as an examination subject in the Faculty of Law, which made it somewhat inaccessible to history students. The chair in economics, including fiscal law, at Uppsala University was held by David Davidson.12 He had in 1899, in response to the increasing interest in economics outside the law faculty, set out as initiator and editor of Ekonomisk Tidskrift, Sweden’s only professional journal in economics. The reason economics had become so attractive outside the faculty of law, especially among history students, was generally that this subject seemed highly relevant to many of the social issues that agitated politically concerned academicians at the turn of the century. Interest in the subject was at the time enhanced by the presence in Uppsala of Knut Wicksell,13 who was ‘‘a celebrity of ill repute’’ in respectable circles after his many ‘‘scandalous’’ lectures and writings. Wicksell, now in his late forties, had in the 1890s made his fundamental theoretical contributions to economics and was then studying for the law degree, which university regulations required before he could be accepted to the position of docent. However, despite the general interest in economics, very few students outside the faculty of law enrolled for that subject. No doubt picking economics was considered a bit adventurous for a history student who would meet with a different tradition in terms of teaching and examinations than he was used to. Studies in history were centered on participation in a seminar with presentation of assigned works as more important than the examinations of course literature and checks on knowledge from attendance at lectures. In law studies the reading of course literature constituted the main workload for students who generally had little contact with the professors in the field. But a further reason for hesitation in committing study time to economics was apparently Davidson himself, whose personality made him notorious as a teacher in failing to stimulate students and who was furthermore considered somewhat of an examination risk factor. In the fall of 1899, Heckscher deplored that he did not have the opportunity to listen to Wicksell, who as a newly appointed docent now taught a well received lecture course. Heckscher added that, alas, he
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
55
had therefore to fall back on Davidson, who was ‘‘almost as boring as his examinations were impossible.’’ However, Heckscher was not a man to take counsel of his fears and, with Hazelius as an informant, he proceeded with economics under Davidson.14 Heckscher’s Political Interest and Work in Political Science In deciding on his study program, Heckscher, while no doubt yielding to his quest for a challenge, was also economizing on his time and effort. As he set out for a goal both boldly direct and rationally circumspect, he had taken advantage of the examination reform of 1891. This made it possible to aim directly for the higher degree of fil. lic. Thus, in opting to skip the lower level degree, fil. kand., he had to study three fields instead of five. The field Heckscher picked, in addition to history and economics, was political science.15 Much of Heckscher’s continued work through 1900 and 1901, in addition to his historical studies, was his seminar work in political science. That subject had no prominent front figure in Uppsala to fascinate Heckscher in the same way as Hja¨rne in history or representing as much pristine authority as Davidson in economics. This may have been one reason Heckscher, as previously noted, spent a semester at Gothenburg University. What seem to have generally attracted Heckscher to political science, as well as to economics and history, were apparently his now growing political concerns as may be inferred from his participation in organized student union and coterie life. Heckscher had in the fall semester of 1898 joined the conservative student society, Heimdal, to which also most of the above-mentioned Hja¨rne group of history students belonged.16 Heimdal became very much Heckscher’s habitat in integrating his many diverse intellectual activities related to the studies for which he had signed up as well as a number of extramural pursuits.17 He was, as the representative of Heimdal, central in the organizing committee for the Nordic meeting of student unions in 1901. As an officer to the meeting he reported thoroughly on its proceedings (Heckscher 1901a;b). It was on political issues that Heckscher, already in his ‘‘freshman’’ year, wrote the first printed items listed in his bibliography (Heckscher 1897). Political issues were also the subject area for his next published writings (Heckscher 1898a;b). These publications dealt with the issue of international arbitration as a way of promoting a more peaceful
56
Theory, Methodology, and History
world. In the fall of 1897 Heckscher had in the main Uppsala daily reviewed a study written by the Hja¨rne disciple Karl Hildebrand (1897), who argued that only a certain type of international conflicts, called conflicts of ‘‘rights,’’ could be resolved through institutions of international arbitration, while so-called conflicts of ‘‘interests’’ could not be resolved through such procedures.18 As regards the latter point, Heckscher argued to the contrary. The next year in the same daily, Heckscher followed up with a discussion of the possibility of abolishing international warfare. He now advanced an argument for the establishment of a progressive global ‘‘legal state’’ (ra¨ttsstat). In his political studies Heckscher interacted with other members of the Hja¨rne group in a deepened focus on the eighteenth century parliamentary proceedings, which were in some degree parallels to the contemporary constitutional issues in Sweden at the tine. Heckscher himself contributed a paper on ‘‘Tillfa¨lliga utskott’’ (occasional committees) dealing with the eighteenth-century organization of work in the Swedish parliament. This paper was subsequently printed (Heckscher 1902) and should be noted as his first published research contribution. The political science paper is of great interest as evidence of Heckscher’s early orientation towards social science as an aberration from the main methodological tenets to which he as member of the Hja¨rne group so far had subscribed. The paper was a product of a rather prolonged seminar participation in the spring of 1901. Heckscher had here been forced to try his hands at some statistical fact gathering and elementary processing in dealing with an historical issue. Generally the stern Rankean principles of source criticism, to which all professional Uppsala historians in those days subscribed, had led most of them to shun ‘‘historical statistics,’’ a term referring to the numerical information generated by the past about itself. In consequence of the source-critical view, most Swedish historians in those days, were apparently negative toward statistical research methods. Getting one’s hands dirty in any type of statistical fact-gathering was not the virtue it is today, but was rather, especially in the high Hja¨rne quarters, frowned upon. Thus, Heckscher had already in his political science work taken a first step on his course of methodological separation away from being purely a political historian. However, the most immediate importance of Heckscher’s work in political science was that it seems to have settled his mind for the choice of licentiate thesis topic in history. Heckscher’s earlier attention
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
57
to arbitration topics, as well as other aspects of political conflict, alerted him to economic policy issues focused in his work on the mercantile system and later central also in his work on the economic history of Sweden. Thus, since eighteenth-century parliamentary life was the theme on which Heckscher had done his most intense research before setting out on his licentiate thesis work, one may rather safely conclude that political science was the main breeding bed for Heckscher’s conception of his thesis topic. Most notably, Heckscher’s research for this paper was a quite integral part of his historical studies at the time when he was already working on his thesis topic. But the question that still needs an answer is to what extent political science contributed to shaping the approach of the licentiate thesis. Studies in Economics Under Davidson Heckscher’s studies in economics are of course a central chapter both in the making of the economic historian and in the shaping of his methodology. Here the Davidson’s impact was a rather complex and intriguing one. Heckscher’s own testimony of the role of Davidson as a teacher of economics was negative. According to Heckscher, Davidson seems not to have been very anxious to increase the enrollment in his subject. He actually appears to have dissuaded Heckscher from attending his lectures. This somewhat shirking attitude to his teaching obligations made him, as noted earlier, problematic to an enthusiast for economics like Heckscher. Also, Davidson was generally too negative and little appreciative in his views on the state of the art in his field. Very few of the received works in economics passed muster under his critical eyes. Furthermore, during the years when Heckscher studied under him, Davidson seems to have become more influenced by the historical school than formerly and did not offer much of the synthesis brought to the fore internationally in neoclassical economics at that time. However, Davidson’s well-known strong inclination toward Ricardo was now beginning to dominate, possibly reinforced by the constructive influence of Wicksell. Heckscher was despite his antipathy to Davidson’s instruction methods impressed by that feature of Davidson’s teaching where Davidson displayed his analytical edge and ability to hold on to long chains of deductive reasoning, a trait Heckscher might have considered an asset also in his own intellectual equipment (Heckscher 1952).
58
Theory, Methodology, and History
But not much is known about Heckscher’s studies under Davidson. Some clue may be obtained from the fact that Heckscher had in a meeting with Davidson in 1901 asked for a reading course that fitted his thesis plans. This indicates that he seems to have had his thesis topic somewhat defined when he set out on his studies in economics. Whether this meant that he had a set of specific questions that economics should help him answer is far from clear, however. Having made economics one of the three fields in his licentiate degree, Heckscher from 1901 on worked closely under Davidson in the later phase of his studies in Uppsala. Highly motivated, Heckscher devoured the learning that Davidson, despite his noninviting lectures, offered in economics. Because the great majority in the audience obtaining Davidson’s instruction were law students who seldom aspired for more than a ‘‘pass’’ in the subject, Heckscher stood out as one of a select few, who seriously pursued higher studies under Davidson at that time. After the departure of Hazelius he may have been the only one and assumed the position as somewhat of an unofficial assistant to Davidson. In that capacity, Heckscher played quite a major role. He carried out what should be considered amanuensis chores for Davidson by setting up the guidelines for studies in economics, first for students in the faculty of law and then also for students in the faculty of philosophy (Heckscher 1904a). Furthermore, toward the end of his degree work he assisted students with examination difficulties, apparently with Davidson’s blessing, by offering private tutoring. A most important event in Heckscher’s studies under Davidson occurred already in 1902. He was after only one year of concentrated work in economics, appointed by Davidson to assist as an official discussant of a doctoral dissertation in the faculty of law.19 The occasion was the assessment of how a an independent-minded law student, Nils Stjernberg, also at the time chairman of the Uppsala Student Union, had handled his subject matter (Stjernberg 1902). Stjernberg, after straying away from the beaten track in his field, had produced a tract with the title ‘‘Till fra˚gan om de s k rent ekonomiska kategorierna’’ (On the so-called purely economic categories).20 Assessing a dissertation on that topic required the expertise knowledge not only in jurisprudence but also in economics. As Heckscher, with the exception for Davidson, was the only one around who was sufficiently versed in economics, he was called upon to serve as the second discussant. Stjernberg’s dissertation was exceptional in many ways. First of all, a doctoral degree was seldom awarded in the Faculty of Law. Secondly,
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
59
dealing with the conceptual issues on the interface between law and economics was a daring enterprise in Uppsala in those days. Heckscher reports, quite stirred, that Stjernberg’s study was met with ridicule in the camp of jurisprudence and with condescension by the economists, meaning Davidson. Unfortunately, what Heckscher said in his capacity as opponent is not known. He had prepared fifteen points, but his discussion was cut short as he ran out of time. This opposition may have been much more important to Heckscher than the passing event Heckscher’s later reporting held it to be. At least some of Heckscher’s points must have dealt with the clash between the conceptualization approach of jurisprudence and the anticonceptualization approach of the historians. The latter had become an issue among Hja¨rne’s disciples following some noted works by Ede´n (1896, 1900) and the previously mentioned Hildebrand (1897). Heckscher’s views on this score had undoubtedly a bearing of central importance for his later plea for theory in economic history. But in contending with the perplex that nomothetical theory posed for the historians, Heckscher had little support from the social sciences he had included in his degree in addition to history. While political science opened up toward statistics, it was still conceptually too underdeveloped to open up for nomothetical theory. Therefore political scientists were almost as negative to covering law tendencies as were the historians. Thus historians, who disregarded that positivistic social science approach, could still feel quite at home in political science. Things were a bit different in economics. That field was more advanced as a nomothetical science, but that feature was not sufficiently pursued and underlined in Davidsons’s presentation. Davidson had once written a dissertation querying about the laws of capital formation and had followed it up with a study of the theory of rent (Davidson 1878, 1880), but he had as much a legal as an economic mind. While the training in jurisprudence sharpened his conceptual powers, the analytic taxonomy of that field was more normative than behavioural. This made his method of reasoning casuistic and little bent on driving points home in a generalising fashion. Furthermore, Davidson had in his young days been much exposed to the teaching of Knies at Heidelberg, who through his abstract, excruciating approach had impugned Davidson’s attitude toward economic theory.21 In his above-mentioned doctoral dissertation of 1878, Davidson had actually been influenced by the Austrian approach, but only more
60
Theory, Methodology, and History
narrowly by Menger. In working out a theory of capital formation he had absorbed the idea of the hierarchy of wants, but without accepting Menger’s related implicit notion of marginalism. Thus, since Davidson had missed out on the emergence of the neoclassical revolution, he was not the one to push Heckscher into the neoclassical camp. Heckscher had his first encounter with neoclassicism, basically as mentioned earlier, through his reading first of Gide (1899) but then especially also Marshall (1890) and Wicksell (1901). Davidson had all these writers on his reading list, but one should note that it was Heckscher who had compiled that reading list. Thus, although Heckscher’s studies in economics at Uppsala may have entailed only a select absorption of economic theory compared to his vast reading of history, they were by no means minor and far from shallow. Heckscher no doubt became rather well groomed in classical economics during his Uppsala years. But he also accepted the criticism expressed by the historical school against ‘‘the whigism’’ of the classical authors. This criticism made clear that classical theory was mainly a specific model, which had an ‘‘institutional fit’’ only to the particular time and place of England in the early Industrial Revolution. More generally, like the historical economists, Heckscher noted the blindness of the classical economists to the fact of institutional evolution in response to technological change. The institutional specificity deprived their theoretical conceptions of any universal applicability. The only general validity of classical theory that Heckscher retained as his ‘‘keep’’ was what might be called ‘‘the economic logic in the small,’’ the type of partial deductive reasoning demonstrated by Ricardo. This was the residue of classical theory that had been transferred into the now emerging neoclassical theory, where it was expanded into a stringent general theory of markets and the market system. Thus, Heckscher never went as far as some of the most extreme members of the historical school, who totally rejected classical theory. Like most members of that school Heckscher only criticized the institutionally specific model set forth by the classical economists. What Heckscher had retained via Davidson was their more rigorous general economic market logic, which made it possible for him to absorb and digest also the neoclassical involution. On that basis Heckscher arrived, through reading and efforts of his own, at what we might today identify as the textbook neoclassical reconstitution of a Millian stationary state, the neoclassical macro system model. This conception served as Heckscher’s point of departure for his later application of
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
61
economic theory in clarifying historical economic policy issues. That Millian basis would, at the side of Heckscher’s later reading of Marshall and Wicksell, account also for the quite Ricardian qualities of his later contribution to international trade theory. That contribution of his appeared properly enough in a festschrift to Davidson (Heckscher 1949b).22 It should also be pointed out that Heckscher, because he had obtained a fundamental grasp of Ricardian deductivism from Davidson, also learned a lot about deductive socialistic thinking from him. Here Heckscher especially absorbed the theory of Davidson’s favorite, Rodbertus. Understanding Rodbertus also made for a fundamental understanding of Marx, whose deductive approach Davidson had penetrated in his above mentioned dissertation. All this made Heckscher well equipped to take on the socialists in his later heated controversy with them (Heckscher 1908b). It may be argued that since Davidson did not give Heckscher a fully satisfactory overview of economics, Heckscher had himself to put together the synthetic whole. However, for this he equipped Heckscher well, by making Heckscher’s grounding in the field conceptually solid. Thus Heckscher’s logical mind could probably perform that task well enough to rescue him from surrendering to the overall antitheoretical teachings of the historical school. But it is important to note, as a major deficiency in the teaching that Heckscher received from Davidson during the Uppsala period that he was inadequately informed about the increasingly predominant neoclassical economic theory. Heckscher’s exposure to Wicksell and to available literature, which must soon have included also Marshall, was probably not enough to make up for these lacunae in his overview of economics. Heckscher probably did not begin to make up for this deficiency until he sat down to write the manifesto. We may, as a summing up of Heckscher’s studies under Davidson note that he became much more knowledgeable in economics than has generally been recognized. We may then conclude that the study of economics as a conceptual approach served as a second source of aberration from the tenets taught by Hja¨rne. But, as with the statistical approach to which he was exposed in political science, the impact of the conceptual approach in economics may not, as we shall note, have had an effect rapid enough to influence his work on the licentiate thesis. The influence of these aberrations from the Ranke school guidelines did not show up until in the manifesto. To see the historical ‘‘musts
62
Theory, Methodology, and History
and must nots,’’ there is a need to take a closer look also at the holds that Hja¨rne’s teaching had on Heckscher. The Grounding of the Historian and the Importance of Hja¨rne While Davidson, despite his cooling influence, may be seen as important for Heckscher’s intellectual development by ushering him into the world of academic economics, he obviously never served as the mentor to Heckscher in the same way as Hja¨rne. The latter seems to have exerted a strong not only professional but also ideological, influence on Heckscher as well as on his other disciples. However, while it seems clear that what we have called the making of the economic historian was a process during which Heckscher gradually emancipated himself from the methodological tenets of Hja¨rne, this did not mean a weakening of the ideological bondage to his mentor. This bondage is of importance in understanding ‘‘the making of the economic historian’’ insofar as it had an at least indirect bearing on Heckscher’s conception of the task of economic history. But the character of this bondage seems too complex for a ready assessment here. More pertinent and less remote to the overall theme of the present essay was the influence that Hja¨rne exerted on the professional level. Abiding by Hja¨rne’s strictures regarding the pursuit of historical craftsmanship on the level of the work floor was for his students almost a matter of obedience to second-nature lifestyle imperatives. The inculcation of the principles of historical source criticism was the mark of a professional historian. Thus the teaching of the techniques and methods of historical research, in the spirit of Ranke, was basic to the training of the historians in the Hja¨rne seminar. As an example of what Heckscher’s work under Hja¨rne on that score involved, may be mentioned the first historical writing that Heckscher got published (Heckscher 1898c). In that paper, titled ‘‘Va˚r a¨ldsta kungaa¨tt’’ (Our oldest royal ancestry line), Heckscher offered an account of the most recent controversy and fate concerning the Yngliga Saga, a tale by the medieval Icelander Snorre Sturlasson. Heckscher could report that as a result of a sharp Rankean scrutiny in the hands of scientific historians, that story about the birth of Sweden as taught to all Swedish schoolchildren in those days, had now to be relinquished as completely a myth. Heckscher praised this result highly, not only because it demonstrated the power of the Rankean source critical approach, but also be-
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
63
cause it fitted his vision of history mediated by Hja¨rne, implying that the emergence of Sweden as a state and nation had to be placed in the wider European context as the rise of the West out of the age of antiquity rather than as a purely Germanic or Nordic genesis.23 This wider perspective on Swedish history, was an essentially ‘‘medievalist’’ vision in the sense that it required an overview extending both forward to the present and backward in time to antiquity. Heckscher’s ‘‘medievalist’’ outlook made him especially receptive to the Hja¨rne message that the historian’s ‘‘ulterior motive’’ always was a concern for the time in which he lived. What Heckscher seems especially to have absorbed in Hja¨rne’s teaching was a type of Whig interpretation of history that can be called ‘‘methodological presentism.’’ Hja¨rne projected a perspective on the past that was implicitly comparative in search for universal features in the historical process implying also a central understanding of the present day and age. Generally, historians, in turning to the past, often quite intuitively choose those periods or problems that seemed to be parallels to the present, but this should be done more consciously in order that the methodological problems of this approach should be clarified. In searching for basic parallels to the present the historian should, because history never fully replicated itself, also be alerted to the contrasts with the present which in that way would be even further illuminated. In his methodological presentism Heckscher, in the writing of the Economic History of Sweden, eventually went even farther than Hja¨rne. This Heckscher’s advance meant that the historian in his searchlight on the past was looking for factors that pointed forward in the historical process and thus ultimately also to the contemporary world to which the historian did his reporting. It meant asking in what way a past was part of the root system for what followed. This deeper Whig interpretation of history involved an effort at perceiving those particularities in the onward frontier that linked up to the present. Aside from informing about the important content of Hja¨rne’s teaching that Heckscher absorbed and extended, the above informs also about the methodological approach that Hja¨rne by example implanted in his students. It makes clear that Hja¨rne turned them into more than Ranke proselytes. He also taught them much about the essentials of synthesis in historical reconstruction. Reconstruction must necessarily be based on interpretation, but interpretation presumes the existence of an object to be interpreted. Thus interpretation is both a matter of analysis of causal links in the nodal core of a historical web and of
64
Theory, Methodology, and History
putting together the larger picture of historical change and context. Heckscher, who in general was Hja¨rne’s foremost follower in emphasizing the need for synthesis, was also on that score holding a view beyond Hja¨rne in pushing for the need to make syntheses as much as possible empirically objective. The reason he did that, or could do that, is to be found in his deviation from Hja¨rne’s anticonceptualist stance. Heckscher tried to escape from Hja¨rne’s reluctance to offer too daring historical interpretations premised on abstract concepts. Hja¨rne’s main ambition, following Ranke, was confining history to the task of establishing ‘‘wie es eigentlich gewesen’’ through resort only to the sourcecritical approach, where all problems of synthesis were to be solved through a purely creative act of overall assessment that could not be anything but subjective. Heckscher held that historical syntheses may, like castles in the sand, be obliterated when there are no supportive conceptual pillars. Heckscher’s explicit advocacy for the necessity of abstract premises—and, we should add, statistically reinforced—for historical syntheses was his basic deviation from Hja¨rne. In Heckscher’s view Hja¨rne’s position precluded the study of social and economic progress as a subdiscipline within the field of history. Actually, Hja¨rne’s embracing the uniqueness of historical events implied a global vision that ruled out a comparative approach as regards the total process. Hja¨rne’s position meant to Heckscher a radical taboo against all attempts to deal with history in theoretical terms.24 However, although Heckscher did not accept Hja¨rne’s anticonceptualism, he agreed in his formative years to Hja¨rne’s opposition to positivism and also to his opposition to nomothetic explanations of historical change. In the early phase of his productive life as an economic historian Heckscher only accepted nomothetical theory insofar as his query about the past was confined to the workings of the system as opposed to its evolution (Heckscher 1922a, 1929, 1930b, 1933, 1936b). But as he in the later phase adopted a statistical outlook and applied a statistical approach, nomothetical theory eventually had to be resorted to even in accounting for economic-historical change (Heckscher 1951a; Henriksson and Lundahl 2003). Thus Hja¨rne long had a confining influence on Heckscher’s methodological position and may be said to have shaped his stance not only as regards the conception of the task for economic history, but also as regards its scope and method. Not until Hja¨rne’s hold on Heckscher’s conception of its task began to ease did Hja¨rne’s confining influence on Heckscher’s views regarding the scope and method of economic history soften.
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
65
But with this we are back facing the issue of Hja¨rne’s ideological importance and we shall therefore on that score only make one point previously touched upon in a footnote. The ideological lead of Hja¨rne implied for Heckscher a moral adherence to a universalizing western European culture perspective, to which Swedish nationalistic concerns were subordinated. This wide perspective was not only the basic framing for Heckscher’s work on mercantilism but also what inspired him in the writing of the Economic History of Sweden. Here Heckscher, influenced by Hja¨rne’s conception of history, actually transcended the narrow economic perspective and opened to the spiritual dimension even of economic life. This strongly permeated Heckscher’s pursuit of economic history from the start. Heckscher’s Hja¨rne-inspired conception of historical craftsmanship as a calling for economic historians in the service of a cultural mission should be kept in mind for understanding ‘‘the making of the economic historian’’ as that process unfolded in Heckscher’s licentiate thesis and subsequently in the methodological manifesto. The Licentiate Thesis: A Specimen for the Historian’s Craft After having spent the first Uppsala years in apprenticeship for the craft of historian, Heckscher finally turned to the task of doing the specimen work required for the fil. lic. degree, which was considered the lower entry level to the profession before the doctorate. In 1903 he submitted a licentiate thesis to Hja¨rne with the title ‘‘Studier till Produktplakatets fo¨rhistoria och historia’’ (Studies in the prehistory and history of the Navigation Act) (Heckscher 1903). This study dealt with the Swedish Navigation Act of 1724, which was a decree favoring Swedish vessels in the import and export of goods to and from the country.25 Not revealed by the overall title, the thesis also entailed, as a framing background, an overview of the evolution of the system of economic policy prevalent in the Western world before the age of economic liberalism called the mercantile system. With this thesis Heckscher initiated what in his vernacular may be referred to as the two ‘‘Faustian’’ ventures in his scholarly life.26 Through the inclusion of the lengthy background survey of the mercantile system, the thesis was first of all the very beginning of his great work on mercantilism, the major study of the history of economic policy that evolved as the central theme in his historical research during the early half of his scholarly life (Heckscher 1935a). Secondly,
66
Theory, Methodology, and History
from the main part of the thesis unfolded his massive work on Swedish economic history, his so-called real task, which became his predominant preoccupation in the later half of his life (Heckscher 1935b; 1936a; 1949a; 1954). The thesis was presented to Hja¨rne as a handwritten manuscript in November 1903. Apparently most of the writing had been done during 1903. Judging from purchasing dates of books preserved in the Heckscher book collection and from library loan records, one may conclude that his research on the framing part dealing with the mercantile system took place mainly in 1902. But preparation for this part, as well as for the rest of the thesis, may have been initiated much earlier. As noted above the idea for the thesis topic in general may have arisen already in 1898 during Heckscher’s study period under Stavenov in Gothenburg. However, the most likely starting year for the substantial work on the project, dealing with the specifics of the Swedish navigation act, is 1901. As was noted above, in that year Heckscher asked Davidson about a reading list that fitted his thesis plans. As only the embryo of his monumental study, Heckscher’s general introductory discourse on the mercantile system of course lacked many features of his later work, but some are already strikingly present in this early presentation. The account had all Heckscher’s general characteristics of writing style and eloquence and, of course, his meticulousness.27 The main difference to his later work was that his presentation was not based on primary sources. It was only a synopsis derived from his reading of the secondary literature. The central authors were Schmoller and Cunningham, but there were of course references also to all other leading commentators on the mercantile system such as Ashley and Hewins. Heckscher even referred to Marshal, but not as approvingly as he did later. In general, he was in many ways critical of the main authorities he built on. In particular, Schmoller was singled out for not drawing sufficiently on primary sources. As regards detailed facts, Heckscher’s general discussion of the mercantile system was in no way original, and even his overall conceptions hinged much on the literature he referred to. But on that level he often made startling remarks and arrived at questions that called for the explorations he was later to undertake. Of course some queries posed in the thesis had later to be discarded, but other observations proved to stand subsequent validation research well. But any applica-
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
67
tions of the economic theory he had learnt from Davidson are strikingly absent. Actually, Heckscher’s early vision of the mercantile system, albeit with major deficiencies, was stated in this first tract with a vigor and clarity that Heckscher was not always to repeat in the later treatise. When his main points were put forth within a more explicit conceptual scheme and in greater detail, their immediate intuitive appeal, that is part of the persuasive qualities of a good interpretation, were somewhat lost. The account in his later presentations often seems stale in comparison with his youthful impressionistic text in the thesis. The second and major part of the thesis, dealing specifically with the Swedish Navigation Act of 1724, was in contrast to the introductory part, a report on archival research covering the background of the act and its antecedents. Here Heckscher’s approach was not very different from what might have been expected from any political historian. Yet, his focus back in time on the antecedents of the act was notably deep even for a historian. Before providing a thorough account of the processing of the act in parliament and of the politics of its implementation, it offered a penetrating overview of earlier Swedish foreign trade and shipping policies since the seventeenth century. The methodologically interesting part of the inquiry, anticipating a central feature of his own later study on the economic history of Sweden, concerned his emergent attention already at this time to the views and understanding of the policy makers and the debaters. Heckscher, in commenting on the deficiency and lack of data at the time on the Swedish commercial fleet, showed how fatally inaccurate their perceptions of the real world could be. It is notable that he constructed a statistical table of information based on primary materials in an attempt to assess the impact of the act, but in this he only succeeded in capturing a short-term partial effect. But Heckscher’s statistical efforts are of course very interesting as evidence of his ambitions, which meant illuminating the wider and long term implications of the act. It would here take us too far to account in more detail for the content of the thesis as we are only looking for evidence that illuminate Heckscher’s progress toward becoming an economic historian and his advance of a proper methodology for that field. On that score, the licentiate thesis showed how Heckscher, schooled as a historian and equipped with the historical method of source criticism, at an early phase of his work tried to explain the development of economic policy
68
Theory, Methodology, and History
with surprising little use of economic theory. We may especially note that he had little to say on the effects of economic policy. He was more interested in its causes. But in his attention more to motives than to contexts, Heckscher in fact pointed to the factor that made the policy efforts ineffective. This opened for his later more systematic observations on the policy makers’ perception of the world. Heckscher already in the thesis extended his coverage of this perception to include economic thinking. It is clear that Heckscher, despite having made advances in resorting both to statistics and to some conceptualizing, was still far from being an economic historian. In his attention to economic policy and thinking, his approach was essentially that of a general political historian. Further, he did not cover the state of the economy in any modern sense, that is, of economic life as the object of economic policy. Although he had studied economic theory this did not show up neither explicitly nor in the structuring of his approach. Thus he had little to say on questions of method in exploring past economic life and its evolution. Heckscher’s main deviation from Hja¨rne so far was his choice of topic. But from what we know, Hja¨rne had raised no objection to Heckscher’s thesis topic and was pleased with the work. A Manifesto for Economic History The licentiate degree Heckscher earned in 1904 was an intermediatelevel degree, which had taken seven years of university studies. Naturally the question whether he should go on for a doctorate or try some alternative career had now to be faced. Applying for a position with the Kommerskollegium (Board of Trade) had loomed as an option at the time of setting out on the thesis work, but now his ambitions were higher. While Heckscher had been far from pleased with his performance on the examinations for the fil. lic. degree, Hja¨rne had thought highly of his thesis of which also Heckscher felt proud. Thus, although Hja¨rne appears not to have much encouraged Heckscher to go on for an academic career, Heckscher was himself mentally set to follow up on the licentiate degree with a doctoral degree. The main factor in the scales was apparently that the financing of further academic studies seemed somewhat uncertain. Heckscher’s parents were probably enough well off to support him through for a further prerequisite period of dissertation research, but quite understandably Heckscher, now twenty-five, wanted to become economi-
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
69
cally independent. However, the uncertainty about the financing of further studies was soon resolved. Heckscher had, some time in the early spring of 1904, been approached by the board of the Swedish state railroads for a coming anniversary study about the economic significance of the development of Sweden’s state railroad system. This was to become the subject for his later doctoral dissertation (Heckscher 1907). Soon after the railroad proposal in early 1904, Gustav Cassel, who had just been appointed to the chair in Economics at Stockholm University, inquired if Heckscher would be willing to do some teaching in that year’s summer courses. This contact with Cassel then led to Heckscher being in the fall of 1904 appointed amanuensis without pay in the Institute of Social Science that Cassel had just initiated at Stockholm University. During the five year sojourn under Cassel (1904– 1909) that followed, Heckscher was completing his dissertation, was appointed associate professor in 1907, and in 1909 was called to a chair in Economics and Statistics at the Stockholm School of Economics. This may seem to have been the beginning of a career quite far from the field of economic history that Heckscher conceived of in 1904. But with the manifesto of 1904, the making of the economic historian had become inevitable as was made evident by his later career. The Presentation of the Field of Economic History It was probably with the noted academic opening at Stockholm University secured, that Heckscher, in anticipation of the methodological issues in his coming dissertation work, in the fall of 1904 submitted the renowned paper to Historisk Tidskrift where, in the manner of a manifesto, he presented economic history as a distinct special field of academic study (Heckscher 1904b). Here, after surveying earlier work in the field, he outlined what he, in somewhat of a personal declaration of intent, considered to be its task, scope, and method. Heckscher’s paper is divided into two main parts. Drawing on his thorough acquaintance with the literature built up during the work on the licentiate thesis, Heckscher offered in part one a comprehensive account of how the field of economic history originated. Although he took pains to note that it had important antecedents also among pure historians dealing with the Middle Ages, his central contention was that it originated mainly in economics. He presented its rise as a criticism of classical economists for ‘‘their total blindness to the historical and evolutionary aspects of economic life.’’ Heckscher described the
70
Theory, Methodology, and History
phases of the development of this criticism in various countries and commented incisively on various schools of thought. In so doing he offered a very interesting account of the early socialist writings pertinent to the theme with attention not only to Marx but also and especially to Rodbertus. However, his main attention was devoted to the historical school. He noted the leading German names of its older generation, where in addition to the previously mentioned Knies he also of course included Roscher and Hildebrand. Yet his focus in the review of the historical school concerned the younger fraction, which Heckscher saw as the real breakthrough for economic history. Here the Germans Schmoller and Wagner received due attention, but Heckscher’s most interesting presentation concerned the English branch. Ashley, Toynbee, and Cunningham were presented with apparent enthusiasm as the central figures. With the mercantilist system on his mind, Heckscher could give a quite enlightened presentation. He of course gave key attention to Cunningham for clarifying broadly the nature of state power and the emphasis on defence in the evolution of mercantilist policies, while Schmoller was noted for his attention to the territorial dimension and the importance of transport and communication costs as a central conditioning factor in that process. It is impossible here to offer a full and fair review of Heckscher’s impressive survey. Although it was polemically peppered it was still authoritative. It was richly sprinkled with incisive comments on details but offered also many startling vistas. One may note as a gem in the account of the writings of the professional historians, Heckscher’s comments on Macaulay. These reveal his appreciation of Macaulay’s colorful descriptions of the state of the English economy at various moments in the past. They would please even schoolchildren for which such pages in a history book were usually pure horror, Heckscher noted. Another interesting part in Heckscher’s broad overview was his commentary on the evolution of the writings in the History of Economic Thought. The Task and Scope of Economic History In the second part of the paper Heckscher set forth his much noted methodological statement of what economic history purported to be. In this were also included didactic declarations about what it did not purport to be. In drawing that latter line he made two important
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
71
demarcations. In response to what he saw as a central message in the somewhat hegemonic Marxist dogma on the socialist side, he pointed out that economic history was not the same as the economic interpretation of all history that he conceived to be the Marxist point of view. He strongly distanced himself from such a one-sided perspective on the totality of history. Further, in an important addition in the final proofs before printing, he criticized the view on economic history stated by Wicksell in his recent inaugural lecture at Lund University (Wicksell 1903). Wicksell, in clarifying his disagreements with the historical school, had argued that economic history was an important empirical base for the development of economic theory insofar as the inductive generalizing from that type of field experiences rested on the interpretation of historical events as ‘‘laboratory’’ outcomes of ‘‘as if’’ staged experimental situation for deductive retrodictive testing of general causal hypotheses. Heckscher offered no compliments to Wicksell for that view on the making of theory. While thanking Wicksell for his supportive attention to economic history, Heckscher instead faulted him for not understanding its scientific task, which according to Heckscher was not at all to develop theory, but rather to explore the development of economic life (Heckscher 1904b; Henriksson 1991). In his reply to Wicksell Heckscher had thus also stated in a more direct way what he considered economic history to be by pointing to its task. The task of studying and explaining economic development was to him entirely different from the task of economics which was to explore the laws regulating the workings of economic life. Heckscher’s made some seemingly rather ambiguous comments on his adherence to the so-called historical school, but his position becomes crystal clear once we recognize that sharp division of labor he imposed on the two fields. In following up on his criticism on Wicksell that the task of economic history was not at all to offer an alternative way to reach a better economic theory, he pointed out that that, unlike the historical school at the time, economic history had no aspirations to replace theoretical economics with historical economics. In his general elaboration in the paper he clarified that the ambition of economic history was not to offer a substitute to economic theory. As an economist Heckscher was the opposite of a proponent of the historical school although he may have remained sceptical of the Wicksellian view. But as an economic historian he saw no inconsistency in accepting the central critical stance of the historical school against economic theory in so far as the task of economic history, as Heckscher pointed out to Wicksell, was to
72
Theory, Methodology, and History
study and explain economic development. However, Heckscher’s position did not rule out the application of economic theory in historical research, but this was confined to clarifying only the workings of past economic life and, to repeat, not to be resorted to in explaining economic evolution. Economics and economic history had entirely different objectives. In stating the above views on the task of economic history, Heckscher made quite a breach with the position that was implicit in his licentiate thesis. The breach was quite thoroughgoing as it entailed a new view of the object or scope for the analysis. In writing the thesis Heckscher had methodologically still been a political historian dealing with a specific instance of economic policy and the related economic thinking. The basic change that had occurred, as recorded in his position of 1904 in comparison with the position in the licentiate thesis, was a shift in scholarly focus that meant the elevation of economic life more narrowly conceived, to center stage of the historical study. Instead of being merely a background in a political account of government economic measures, economic life should in an economichistorical account be its very core part. Of course, in general the scope of economic history would, in addition to its core object, still include both economic polices and economic thinking. But only the core, that is, economic life in a more narrow sense, should, in Heckscher’s view, be the constitutive part of the general object of that discipline. In his more detailed presentation of the conception of economic history to which he had now arrived, Heckscher elaborated on this broad scope of the discipline. He first reverted to the coverage of the notion of ‘‘economic life’’ for a further fine but crucial point. He advanced what in the following might be called a dualistic view on the economy as the object of inquiry. According to this dualistic view, economics and economic history both had the economy, defined as an economic system, as their object of study, thereby sealing a logical tie between these two disciplines. Heckscher strongly underlined this formal identity of the scope of economics and the scope of economic history. To cite from Heckscher 1904b: ‘‘As long as economic life is a closed system, i.e. as long as economics is an independent science—also economic history is a closed entity, but no further.’’28 However, by stating this identity of the formal object of inquiry in the two fields, Heckscher was able to make a very clear demarcation between them as regards their research and study assignments. Despite their common focus, the two fields had entirely different analytic
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
73
and synthetic tasks. While economics had to clarify the workings of economic life; that is, its mechanism of short-term performance, economic history was, as stated in Heckscher’s comment on Wicksell, concerned with the long term evolution of economic life.29 Having set forth the above noted segmented view of the scope of economic history as a field covering not only the history of economic life in the narrow sense referred to as the core, but also the historical realms of economic policy and economic thinking, Heckscher had, as a major issue for analysis, raised the question about the interrelationship between the components in the triad. His views on the interdependencies between the core and the two other components of economic life may be summed up in a number of quotes. A first point concerned their relationship in general. Heckscher wrote: ‘‘These three parts of total economic life were of course in many ways interrelated, but each of them was at the same time influenced by its own specific factors and forces that were separate from the influences affecting the other parts’’ (p. 186). He then attended separately to each of the components. The first point now concerned the above noted focus on the core object of economic history. He clarified what the coverage of the core entailed; that is, what the study of past economic life and its evolution implied. In his words it ‘‘concerned the different branches of productive life, such as the history of agriculture from the times of Germanic extensive husbandry to modern intensive forms of land cultivation, the history of industrial life from domestic production to handicraft and from cottage industry to manufacturing, the transition from natural economy to a money economy, the history of credit and capital etc’’ (p. 185). To that he added the following important rider: ‘‘On the basis of such accounts the economic historian then has to construe a picture of the living conditions of the different classes of people’’ (p. 185). Turning to economic policies and economic thinking, as subsidiary elements of economic life, Heckscher argued, what he had already noted in the licentiate thesis, that they evolved only tenuously related to the hard facts of economic life i e to the core. Economic policies had according to him for ages been determined ‘‘more by dynastic and military interests than by economic considerations’’ (p. 186), and as regards economic thinking his view was that it had been ‘‘similarly dominated by the old hegemonic moral and philosophical systems of a period’’ (p. 186).
74
Theory, Methodology, and History
These points on the relationship between the core of economic history and its overlay of economic policy and thinking, Heckscher no doubt had derived from his studies of medieval economic life. But while his formulations leave open the possibility, that economic policies and thinking during the following period of the mercantile system were more conditioned by economic needs and constraints than Heckscher believed, he appears generally to have held that this medieval character of economic policy and thought prevailed strongly even in that later period. But as regards economic policies in the modern age, Heckscher seems to have argued that a central task for the economic historian was to find out how and to what extent economic conditions and considerations influenced economic policy decisions. This followed from the duty of an economic historian, who as the general rule ‘‘had to show how the representatives of society from the village commune to the state conceived of their task in relation to economic life; what they fought against, what they promoted and what they left untouched’’ (p. 185). Heckscher also made the point that, as a general rule, the task of the economic historian with respect to economic thinking; that is, the history of economic ideology and of economic science, was to show ‘‘how the conception of economic life has evolved; what in different ages was conceived as its functions and forms’’ (p. 186). While it is clear that Heckscher’s segmented view of economic life was present already in his thesis, it was in the manifesto underscored as fundamental in marking out what was to be analysed as the core of the object. As the most central point relating to the mercantile system, Heckscher now held that there had been no close connection between the development of the state of the economy in the narrow core sense and economic policy and economic thinking. As regards the influence of economic policy and economic thinking on early ‘‘capitalism,’’ a term for that narrow core conception of the economy that he had not yet discarded, Heckscher argued that the authorities lacked the administrative muscles to achieve anything whatsoever, that any measures were just ‘‘thrusts in the air.’’ But he also noted the role of information failures making the policy efforts have unexpected outcomes and side effects that were generally negative. He also recognized that, as much as the information failure was due to deficient statistical and other factual information, it was also due to inadequate theoretical understanding of the economy.
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
75
As regards the reverse influence of the economic core on economic policy working, through the understanding that would be generated among the politicians through the confrontations with facts, Heckscher of course admitted the possibility that the advance of economic thinking might evolve out of repeated experiences. But here he held a very pessimistic view, as experiences may not have been fully perceived due to the unreliability of or the lack of statistical information, which of course also slowed the advance of theoretical understanding. Interestingly enough, Heckscher did not much elaborate on the possibility of a divergence between developments in economic thinking and developments of economic policy. But such a divergence seems to be implied in the vision set forth already in the licentiate thesis about economic liberalism being the executioner of what mercantilism wanted to achieve. However, he did not offer a major historical interpretation on the premise of a segmentation between economic policies and economic thinking. Such a divergence could perhaps not arise until the emergence of economic science as distinct from the general popular understanding and lay discourse on economic policy issues. It would only emerge as a point in Heckscher’s later writings dealing with the modern phase of economic history. The Methods of Economic History Recognizing Heckscher’s segmentation of the object of economic history is fundamental for understanding his views on methods. As long as the object of study was past economic policies and economic thinking, Heckscher had little reason to reconsider the established approach of general history. Accordingly he said little on the problems of method they raised.30 In contrast, he had much to say on the methods of studying the evolution of the core of economic life in the narrow sense. This requires a number of comments. Heckscher of course recognized that the study of the evolution of economic life in the narrow core sense of the term required methods that are quite different from the methods used in the study of past economic policies and thinking. However, it is important to note, as a first framing point, that Heckscher in fact betrayed a disregard also for methodology. In this he showed the usual impatience and disdain of the creative mind for pedagogic didactics. Heckscher always underscored the need to see issues from a practical point of view, although he usually added that the most practical thing is a good theory. With
76
Theory, Methodology, and History
Heckscher’s adverse view to formal methodological discourse went a similar lack of appreciation of the kind of assistance he might have obtained from the professional philosophers at the time. He actually betrayed a clear dislike of philosophy. The negative attitude to methodology and philosophy was probably an inculcation from Hja¨rne.31 Heckscher appears as yet not to have been especially well read in philosophy, and there is not much evidence from which to infer about where he stood as regards the different philosophies that might be invoked when stating a methodological position in economic history. Therefore it seems somewhat meaningless to query into which philosophical camp he belonged. Quite clearly he was not a utilitarian like Wicksell, nor can he be called a positivist. As an adherent to the Ranke-Hja¨rne view his philosophy of history, if any, was more in the neo-Kantian idealistic tradition. Of particular interest as regards Heckscher’s historical approach as an empirical method is his view on so-called historical statistics. Here Heckscher still advanced the position noted earlier, which was in the main the one common to all historians at the time. He devoted much space in the paper to a commentary on the use of statistics in economic history. Here he could draw on his own earlier experiences from statistical work in political science. These views seem to have been little changed by his experience of writing the licentiate thesis. It is of course not surprising that he made no substantial advance in statistics even in his thesis work, considering that economics was taught in the law faculty. As an historian Heckscher had previously not expressed much thought on questions of principle regarding the handling of primary statistical material for historical reconstruction of economic life narrowly conceived, because he had little need for it. He had been trained to disregard this issue. As long as his main purpose was confined to reconstruct and explain the history of economic policy, as in the licentiate thesis, he had not seen any need for such research. When he wrote his 1904 paper he still had not done any such research. But, with the dissertation project soon engaging him, he had at least responded to the necessity to think about what it meant concretely. Concluding Summary The story told above about the making of the economic historian and the shaping of his methodology underscores that this was a process of
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
77
gradual transformation from a starting position that might be called that of a pure political historian over an intermediate transient phase before finally arriving at the position of 1904. The account has sought to pinpoint the factors that made for the slowness of this process and the notable resilience of his first views. The inculcation into the mores of a general political historian meant a crucial conditioning of the subsequent shaping of his methodology as an economic historian. The creed of the general and political historian in the Hja¨rne tradition meant one ‘‘must’’ and two ‘‘do nots.’’ The ‘‘must’’ meant adopting the source-critical method of fact gathering as the main method in writing history. The first ‘‘do not’’ prevented the use of generic ‘‘historical statistics’’ and the second prevented the use of ‘‘abstract concepts.’’ The two ‘‘do nots’’ were hurdles for Heckscher to overcome before he arrived at his research program. Although Heckscher’s position meant a breakaway from Hja¨rne’s hold on him, insofar as he violated the two ‘‘do nots’’ he still accepted the Hja¨rne ‘‘must.’’ In that way Heckscher as an economic historian never cut off from history completely. In reaching his research program for economic history, Heckscher still accepted the ‘‘must’’ in the historical research program, but refuted the two ‘‘do nots.’’ Instead, he now pleaded for both a statistical and conceptual approach. However, the Heckscher program of 1904 had no use for nomothetical theory in pursuing the main task of explaining economic development. Only the exploration of the workings of the old economic system at a given point in time would call for nomothetical economic theory. Only in that latter case can the Heckscher program be seen as precursor to the cliometric program. That program does not include the source-critical ‘‘must’’ of the history program, and while not refuting the two ‘‘do nots,’’ it is one level higher up by pleading for the use of fully specified models, which presumes a resort also to nomothetical theory and empirical procedures grounded in statistical theory. Heckscher’s research program for economic history comprising tenets about the task, scope, and method of the field may be summed up as follows. Heckscher strongly underscored that the scope of economic history has to be exactly the same as the scope of economics, ‘‘no more, no less.’’ What fundamentally separates the two fields, he declared, are their tasks. While the prime task of economic history is to study and explain the evolution of economic life, the prime task of economics is to explore and explain its workings. The economic historian
78
Theory, Methodology, and History
was by Heckscher here assigned the task of studying economic development apparently as a long term process, while the allotment to the economist was by Heckscher conceived as the study of the short-term behavior of the economic system, its ‘‘mechanism’’ as he called it. Turning then to the point concerning the method of economic history, Heckscher made a declaration that in part followed rather neatly from his declarations about its scope and task. Insofar as economic history was only concerned with the task of studying and explaining economic development as a unique historical process there was no use of economic theory, or for that matter, of any other type of nomothetical social theory. The method called for was simply the method of historians as worked out especially by the Ranke school, namely the principles of source criticism in the assessment of factual events and their causal origins. According to the pure Ranke school, to which Heckscher’s major mentor Hja¨rne belonged, there was no further need for either the conceptual or the statistical approach. But on this point, Heckscher differed from Hja¨rne. Heckscher recognized the need for labels such as mercantilism and the Industrial Revolution in the same manner as the general historians had names for broad phenomena such as the Renaissance and the Reformation. But the strictures of the historical method accepted by Heckscher did not allow abstract labeling of phenomena on the historical time/space coordinates in the manner of the positivists, who urged the study of history for finding the laws of historical development and hoped that the study of the past would yield normative lessons. However, the 1904 statement was formally too narrowly confined in the conception of the task. There was one notable lapse in the 1904 declaration seen as the formal tenets of a research program. In the manifesto Heckscher visualised that the tasks of economics and economic history were neatly complementary. But as soon as he set out on ‘‘doing’’ economic history, he recognized that his declaration had missed the area of overlap comprising the workings of the economic system in the past. This too, of course, was a study that in practice had to be assigned to the economic historian. Regardless of whether such a study was conducted as part of a project with the ulterior purpose of explaining economic development or not, the study of the workings of economic life in the past called for economic theory as much as the study of the workings of economic life in the present. It was for that endeavor that Heckscher later made his well-known plea for theory in economic history.
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
79
This did not mean, however, that he retracted from the core position of the manifesto. To repeat: Heckscher’s famous plea for theory in economic history only concerned the explanation of a repetitive state in the past. The manifesto, as well as all later restatements of the method of economic history, gave economic theory little role as a tool in the explanation of economic development. Acknowledgments The essay builds on research in progress on an intellectual biography of Heckscher. That project was initiated in the 1970s with benign encouragement from his son, the late Gunnar Heckscher. For the present essay I am grateful to Ivar Heckscher and the other Heckscher grandchildren, Eva, Einar, and Sten Heckscher, for having entrusted me with some of the literary remains of Eli Heckscher. Earlier reports of this project are Henriksson (1979, 1987a, 1987b, 1989, 1990, 1991a, 1991b, 2001, 2002a, 2002b, 2003). See also Henriksson and Lundahl (2001, 2003). This essay draws largely on Heckscher’s correspondence, which is available in the Heckscher collection at the Royal Library in Stockholm. The research has been financed by the Jacob Wallenberg Fund. Notes 1. Very few of Heckscher’s methodological writings are available in English, and these cover only his position for a brief ten-year period of the interwar decades (Heckscher 1929, 1930a, 1933, 1939). The first two of these are reprinted in Lane and Riesmera (1953). 2. That task should claim priority in the much neglected field ‘‘the history of economic history.’’ Here an intellectual biography of Eli Heckscher is still missing, especially in view of the attention that in recent years has been devoted mainly to Heckscher’s participation in the public economic policy discourse. For the important work in this latter area, see Carlson (1994). Some work on Heckscher as an economic historian was, however, done by Utterstro¨m (1982) and later also by Magnusson (1991). 3. In addition to his writings in English, mentioned in note 1, the list of Heckscher’s most important writings in Swedish include Heckscher (1904b, 1908c, 1920, 1922a, 1930b, 1936b, 1941a, 1947, 1948) and Heckscher 1951a and 1951b. The last two items are not listed in the Heckscher bibliography (Ekonomisk-historiska institutet 1950) that was published two years before he died in 1952, A selection of Heckscher’s methodological writings in Swedish together with a commentary has recently been reprinted in Henriksson and Lundahl (2003). 4. My view, as previously expressed in Henriksson (1979, 1990, 1991), that Heckscher’s 1904 statement is to be considered a manifesto, has been criticized by C-A Ohlsson (1992). As the issues he raised require an extensive discussion beyond the purview of the
80
Theory, Methodology, and History
present essay, his points will not be specifically commented on in this presentation. However, the approach in the present essay may be seen as an indirect response to his position. 5. This seems to have been a response mainly to pressures from his somewhat dominant mother, Rosa Heckscher (1856–1944). Heckscher’s father, Isidor Heckscher (1848–1923) was probably of a different view concerning his son’s studies. 6. H. Hja¨rne (1848–1922) was professor in history at Uppsala University through 1885– 1913. The summer course in Uppsala in 1895 dealt with constitutional issues in Sweden during the seventeenth century. Thus it is clear that Heckscher, even before he entered Uppsala University, was quite set on studying history. In high school, Heckscher, as his examination paper in Swedish, had written about Robespierre. Only a few formal spelling and grammatical errors seem to have prevented the grading teacher from giving him the highest mark. The teacher informed Heckscher that the paper was stylistically a very impressive achievement. 7. The basic explanation for that transfer was undoubtedly the appointment of Heckscher’s father as Denmark’s consul to Sweden with office in Gothenburg. It led to the family’s moving there from Stockholm. Concern for Heckscher’s study expenses may here have been the primary reason, but Gothenburg University, as a newly opened university in 1891, with examination rights from 1893, may also have offered some purely academic attractions to Heckscher. 8. Ludvig Stavenov (1864–1950) was professor in history at Gothenburg University (1895–1913) and at Uppsala University (1914–1929). Because Stavenow had earned his doctorate in history in Uppsala in 1890, he, too, was at least formally a disciple of Hja¨rne. 9. It provided the nucleus for Heckscher’s build up of his wider social network that evolved from the ‘‘eating team,’’ the so-called matlag, he was invited to join. The social organizing of student daily food consumption into rather fixed table groups at local restaurants had become an institution of student living in Uppsala. Becoming a member in such a group, and in the right kind of group, was in many ways a crucial part of an Uppsala study period for a student. In such a group Heckscher made the closer acquaintances that were to channel his studies. 10. Gunnar Hazelius (1874–1905) was called on the death of his father in 1901 from further academic work to take over the management of the Nordic Museum and Skansen in Stockholm, the great cultural and natural history bequest of his father. However, Hazelius appears to have remained a summoning name for Heckscher and the student circle in Uppsala even after his departure. After the premature death of Hazelius in 1905, this absentee role seems to have elevated Hazelius into somewhat of a symbolic cult figure or icon for an inner circle of his network, which in that phase emerged as a social compact of close friends somewhat later called the Junta. Heckscher was a key member of the Junta and instrumental in seeing the doctoral dissertation of Hazelius through to a posthumous publication (Hazelius 1906). 11. This translation was a main addition to a rather meagre economic literature available in Swedish. There was before Gide no modern Swedish text. Gide’s text did not especially advance the outlook of the historical school, nor was it partisan with the Austrian view or any other specific school of neoclassicism. It is of some note that Heckscher’s first textbook was not Marshall’s Principles, which he later referred to as his early beacon of economic light (Heckscher 1935). This is not denying the importance of Marshall, but the impact of Marshall on Heckscher appears to have been of a somewhat later date. In addi-
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
81
tion to Gide one may also note the work by Nicholson (1893, 1897, 1901) as an early text of some importance for Heckscher. 12. Davidson (1854–1942) held the chair from 1880 to 1919. His dissertation dealt with the economic laws of capital formation (Davidson 1878). A second publication dealing with the history of rent theory was written after a sojourn in Heidelberg under Knies (Davidson 1880). For an understanding of Davidson’s influence on Heckscher, one must also recognize the role of Wicksell as important for Davidson’s thinking from the 1890s and on. For Heckscher’s account of his dependence on Davidson, see Heckscher (1952). 13. Wicksell (1851–1926) was professor in Lund (1901–16). Wicksell had after his doctoral dissertation in 1895 been denied a docentship in the faculty of philosophy, but had been advised by Davidson that he would qualify for a docentship in the faculty of law if he first passed the basic law degree, the jur. kand. (Ga˚rdlund 1958). 14. In retrospect, Heckscher’s apprehensions must have seemed to him quite justified. After Heckscher had begun work under Davidson in 1901, Heckscher’s father intervened on behalf of his son by pointing out to Davidson that the course composed for Heckscher was a bit too substantial. However, the only result was that Davidson refrained from adding any works to the list. 15. Heckscher made a similar decision later in life, when at age forty he finally assessed the possibility of carrying out his long planned ‘‘real task’’ of writing the Economic History of Sweden. After considering the difficulties, he first concluded that the project was impossible, but then valiantly went ahead with it. Although Heckscher always had everything carefully planned it might appear that he also knew how to make intellectual life a perpetual adventure. But what most spurred him from not spending his life in ‘‘vegetating’’ security was not as much a quest for challenge as the moral obligation under the pledge in his Latin ex libris: Non propter vitam vivendi perdere causas (modified from Juvenal’s Satires (8.33)), which in free translation, admonishes a person against indulgent living that may cause him to lose touch with life’s deeper meaning. 16. As a follow-up to his entry to Heimdal, Heckscher became in 1899 a founding member of Nordiska Fo¨reningen (the Nordic Society) that signaled the return of Scandinavianism, the movement that had died in the 1860s when Denmark was let down in the conflict with Prussia on the Schleswig issue. The new Scandinavianism at the turn of the century was, however, a more cultural than political movement. What could be called common Nordic issues were always to remain matters of close concern for Heckscher. More than any other of the Swedish historians and economists, his network extended into the other Nordic countries, Denmark, Norway, and Finland. 17. His general position in Heimdal was that of a librarian combined with a position in charge of some university extension teaching and lecturing services for non-academicians and working people. Generally Heckscher, here appears to have been engaged in Hja¨rne’s political program, ‘‘defense and reforms.’’ Following Hja¨rne, Heckscher supported the suffrage movement as a rational conservative stance for gaining the support of labor on the issue of general conscription, which had been proposed as an enhanced measure of national defence. But it was probably also as a Heimdal representative that Heckscher on February 3, 1902, lectured in Ga¨vle on labor protection laws. This appears to have been his first extramural public lecture on a central issue of the day. 18. Karl Hildebrand (1870–1952) was one of the earliest to influence Heckscher through his writings (Hildebrand 1897a). Among Hja¨rne’s disciples, he was also among the earliest to review writings in the field of economic history (Hildebrand 1897b). He was also
82
Theory, Methodology, and History
a member of the aforementioned junta and remained long politically influential, especially as editor of the conservative daily, Stockholms Dagblad, which up to the 1920s was a major outlet for Heckscher’s contributions to the daily press. He was finally appointed chief of Sweden’s public debt office. 19. As evidence of Heckscher’s early special position under Davidson, one may note that Davidson on that occasion dropped titles with Heckscher. This was a formal act signifying collegiality. 20. Stjernberg (1873–1943) was, in the build-up of Stockholm University, called to the first chair in its Faculty of Law when it was established in 1907. It is of some interest to note that Stjernberg in turn served as a discussant when Heckscher in 1907 presented his doctoral dissertation in history. 21. The position of Knies as a member of the historical school is intriguing. According to Heckscher (1952), Knies cannot really be called an historical economist. In ‘‘bony abstractions’’ he was little different from the classical or neoclassical theorists. This may go far in explaining also Davidson’s somewhat unclear attitude to received theory. In pointing to Davidson’s dependence on Knies and what he may have imparted to Heckscher there is reason to ponder as a memento the fact that Knies had also influenced both Schmoller and J. B. Clark. 22. In his final testimony about Davidson, Heckscher (1952) ranked him as the sharpest in his generation of Swedish economists. What Heckscher referred to in that assessment was the performance stemming from Davidson’s mastery of classical economic theory and especially his ability to apply Ricardian thinking not only to problems in the present but to issues in economic history as well. Admittedly, Davidson did not do much historical writing, but he made major contributions to the history of Sweden’s central bank, most notably his publication in the anniversary series (Davidson 1931). 23. Before he completed his magnum opus, The Economic History of Sweden, Heckscher wrote his interim synoptic overview Svenskt arbete och liv (Swedish Labor and Life) in 1941, which was later translated into English as An Economic History of Sweden (Heckscher 1941b and 1954). It was a work of patriotism at the time when Sweden was virtually besieged by the German avalanche in northern Europe. But Heckscher held his Swedish identity only in the second place. He declared himself to be first of all a citizen of the world and notably allotted his Jewish descent only the third place in such a ranking (Henriksson 1991). For an overview of Heckscher’s position on Zionism and the related issues of what Jewish culture and genes meant to him, see Henriksson (1979). 24. Hja¨rne was of course opposed to the positivism that toward the end of the nineteenth century turned many historians into following Lamprecht’s search for the sociological laws of history in an evolutionary (Darwinian?) direction. That approach had been discussed by Ede´n (1896, 1900). In these writings Ede´n may be seen to have expressed the majority view of the Hja¨rne disciples at the time when Heckscher joined the group. In 1903 Ede´n (1871–1945) was appointed the second professor in history in Uppsala and was thus, together with Stavenow, who succeeded Hja¨rne himself in 1911, the caretaker of the hold of the Hja¨rne school in Uppsala. 25. The second part of the thesis, omitting the background part dealing with international mercantilism, was published (with little revision) in the Hja¨rne festschrift in 1908 (Heckscher 1908a). The first part of the thesis dealing with the mercantile system was never published. However, a very good summary was made available as an encyclope-
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
83
dia article in 1913 (Heckscher 1913). That summary goes to show that Heckscher had as late as 1913 done little further work on the theme of the mercantile system. 26. Heckscher named his licentiate thesis ‘‘Urfaust,’’ referring, no doubt, to the fact that Goethe’s Faust was a life project in which the second part appeared many years after the first part. 27. One may note his painstaking attention to the bibliographical part, which was a very impressive list of readings and references to sources. He was himself aware of the somewhat notable length of the list and made a remark about that in the preface. 28. The Swedish original text runs, ‘‘Sa˚ la˚ngt som det ekonomiska livet a¨r na˚got i sig sja¨lvt slutet—alltsa˚ lika la˚ngt som nationalekonomin a¨r en sja¨lvsta¨ndig vetenskap—a¨r a¨ven den ekonomiska historian ett avslutat helt, men ej la¨ngre’’ (Heckscher 1904b, p. 186). 29. One may here note, as a somewhat startling and seemingly paradoxical consequence of Heckscher’s demarcation line, that the workings of economic systems in the past must logically be seen as the domain of the applied economist rather than the economic historian. In responding to possible objections he could of course retort that what this meant was simply that the practicing economic historian had also to acquire competence as an economist. However, Heckscher did not make that obvious point until the 1920s, when he advanced his much-noted ‘‘plea for theory’’ in economic history (Heckscher 1929). At the time of writing, Heckscher seems to have been so focused on this dynamic task of economic history, and perhaps so blinded by his logic, that he did not realize how seriously he disregarded the intellectual property rights of economic historians. He forgot the rather trivial ‘‘must’’ for economic historians, that they must of course—and perhaps above all—be concerned also with the past economic life as an equilibrating entity in itself; that is, as conceived by economists. The explanation of Heckscher’s lapse is of course partly that, so far, he had not done any empirical historical research beyond the explorations of the economic policies and thinking of the mercantile system. These explorations had not required that he delve into the actual state and development of economic life conceived in the more detailed narrow sense of the activities of consumption, production, and trading. Heckscher was not to be confronted with the need to repair for the lapse until after World War I, when he finally resolved to start work on this real task, the writing to the Economic History of Sweden. 30. One reason Heckscher has so little to say on the history of economic policy is that most of the economic history research he reviewed in the paper had dealt with the Middle Ages. In that era economic policy was mainly undertaken by such powers in economic life as the church and the city-states, while territorial dynastic authorities were administratively weak as public policy organ insofar as they were feudally decentralized. 31. In the introduction to his paper Heckscher presents his own account as only an ‘‘unphilosophical little’’ preview of the field while awaiting an apparently more authoritative account to come from a senior colleague of his. Needless to say, Heckscher’s account preempted the need for such a further contribution.
References Ashley, W. J. 1889–93. An Introduction to English Economic History and Theory. London: Longmans & Co. ———. 1900. Surveys Historic and Economic. London: Longmans & Co.
84
Theory, Methodology, and History
Carlson, B. 1994. The State as a Monster: Gustav Cassel and Eli Heckscher on the Role and Growth of the State. Lanham, MD: University Press of America. Carr, R. 1950. Review of Heckscher, E. F. (1949). The Economic History Review, n.s., 3:246– 249. Coats, A. W. 1957. ‘‘In Defence of Heckscher and the Idea of Mercantilism.’’ Scandinavian Economic History Review 5:173–187. Coleman, D. C. 1957. ‘‘Eli Heckscher and the Idea of Mercantilism.’’ Scandinavian Economic History Review 5:1–25. Cunningham, W. 1892. Growth of English Industry and Commerce in Modern Times. 2nd ed. Cambridge: Cambridge University Press. ———. 1896. Growth of English Industry and Commerce During the Early and Middle Ages. 3rd ed. Cambridge: Cambridge University Press. ———. 1898–1900. Western Civilization in its Economic Aspects, Ancient Times; Medieval and Modern Times. Cambridge: Cambridge University Press. Davidson, D. 1878. Bidrag till la¨ran om de ekonomiska lagarna fo¨r kapitalbildningen. Uppsala: Lundequistska bokhandeln. ———. 1880. Bidrag till jordra¨nteteorins historia. Uppsala: Esaias Edquists. ———. 1931. Sveriges Riksbank 1834–1860. Stockholm: P. A. Norstedt. Ede´n, N. 1896. ‘‘Ett nytt program fo¨r den historiska vetenskapen.’’ Historisk Tidskrift 16:321–331. ———. 1900. ‘‘Fra˚gan om en ny historisk metod.’’ Historisk Tidskrift 20:205–248. Ekonomisk-historiska institutet. 1950. Eli F Heckschers bibliografi 1897–1949. Stockholm: Albert Bonniers. Elvander, N. 1961. Harald Hja¨rne och konservatismen. Stockholm: Almqvist & Wiksell. Fogel, R. W. 1965. ‘‘The Reunification of Economic History with Economic Theory.’’ American Economic Review 40:93–98. Ga˚rdlund, T. 1958. The Life of Knut Wicksell. Stockholm: Almqvist & Wiksell. Gerschenkron, A. 1954. ‘‘Acknowledgments’’ and ‘‘Preface.’’ In E. F. Heckscher An Economic History of Sweden, v–vii, xiii–xlii. Cambridge, MA: Harvard University Press. Gide, C. 1899. Nationalekonomins Grunddrag. Helsingfors: G. V. Edlund. Hazelius, G. 1906. Om hantverksa¨mbetena under medeltiden. Stockholm: Nordiska Museet. Heckscher, E. F. 1987. ‘‘Skiljedom och allma¨n fred.’’ Uppsala Nya Tidning 8:12. ———. 1898a. ‘‘Fredsfra˚gan och historian.’’ Uppsala Nya Tidning 16:5. ———. 1898b. ‘‘Fo¨rsvaret och kriget.’’ Uppsala Nya Tidning 18:5. ———. 1898c. ‘‘Va˚r a¨ldsta kungaa¨tt’’. Fo¨reningen Heimdals uppsatser 4. ———. 1901a. ‘‘Studenternas folkbildningsmo¨te i Upsala.’’ Nordisk Universitetstidskrift. pp. 157–161.
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
85
———. 1901b. ‘‘Studenternas nordiska folkbildningsmo¨te i Upsala. Na˚gra intryck och anma¨rkningar.’’ Nordisk Universitetstidskrift. pp. 238–249. ———. 1902. ‘‘Tillfa¨lliga utskott.’’ Statsvetenskaplig Tidskrift 4:32–59. ———. 1903a. ‘‘Grunddragen av merkantilsystemet i 16 och 17 a˚rhundradena, sa¨rskilt med ha¨nsyn till den yttre handelspolitiken.’’ Unpublished MS. ———. 1904a. ‘‘Nationalekonomi.’’ Studiehandbok fo¨r examina vid Uppsala universitet. Uppsala: Fo¨reningen Verdandi. ———. 1904b. ‘‘Ekonomisk historia: na˚gra antydningar.’’ Historisk Tidskrift 24:167–198. ———. 1905. ‘‘Statistik och ekonomisk historia.’’ Historisk Tidskrift 25:104–110. ———. 1907. Till belysning af ja¨rnva¨garnas betydelse fo¨r Sveriges ekonomiska utveckling. Stockholm: Centraltryckeriet. ———. 1908a. ‘‘Produktplakatet och dess fo¨rutsa¨ttningar. Bidrag till merkantilsystemets historia i Sverige.’’ In Historiska studier tilla¨gnade professor Harald Hja¨rne pa˚ hans sextioa˚rsdag den 2 maj 1908. Uppsala and Stockholm: Almqvist and Wiksell. ———. 1908b. Socialismens grundvalar. Stockholm: Foreningen studenter och arbetare. ———. 1908c. ‘‘Prishistoriska metodfra˚gor.’’ Statsvetenskaplig Tidskrift 10:161–171. ———. 1913. ‘‘Merkantilsystemet.’’ i Nordisk Familjebok 18:174–177. ———. 1920. ‘‘Historia och nationalekonomi.’’ Historisk Tidskrift 40:1–22. ———. 1921. Gammal och Ny Ekonomisk Liberalism. Stockholm: Norstedts. ———. 1922a. ‘‘Ekonomi och historia.’’ In Ekonomi och historia. Stockholm: Albert Bonniers Fo¨rlag. ———. 1922b. The Continental System: An Economic Interpretation. Oxford: Clarendon Press. ———. 1929. ‘‘A Plea for Theory in Economic History.’’ Economic Journal 39, Historical Supplement 4:523–554. ———. 1930a. ‘‘Monetary History from 1914–1925 in its Relations to Foreign Trade and Shipping.’’ In K. Bergendal et al., eds., Sweden, Norway, Denmark and Iceland in the World War, Part III, 127–268. New Haven, CT: Carnegie Endowment for International Peace. ———. 1930b. ‘‘Den ekonomiska historiens aspekter.’’ Historisk Tidskrift 50:1–85. ———. 1930c. ‘‘Natural and Money Economy as Illustrated from Swedish History in the Sixteenth Century.’’ Journal of Economic and Business History 3:1–29. ———. 1933. ‘‘The Aspects of Economic History.’’ In Economic Essays in Honour of Gustav Cassel, 705–720. London: George Allen & Unwin. ———. 1935a. Mercantilism, vols. 1–2. Trans. Mendel Shapiro. London: Allen & Unwin. ———. 1935b. Sveriges ekonomiska historia fra˚n Gustav Vasa. Medeltidshusha˚llningens organisering Fo¨rsta boken 1520–1600. Stockholm: Albert Bonniers Fo¨rlag. ———. 1936a. Sveriges ekonomiska historia fra˚n Gustav Vasa. Husha˚llningen under internationell pa˚verkan Andra boken 1600–1720. Stockholm: Albert Bonniers Fo¨rlag.
86
Theory, Methodology, and History
———. 1936b. ‘‘Den ekonomiska historiens aspekter.’’ In Ekonomisk-historiska studier. Stockholm: Albert Bonniers Fo¨rlag. ———. 1937. ‘‘Materialistisk och annan historieuppfattning.’’ Svensk Tidskrift 27:109–120. ———. 1939. ‘‘Quantitative Measurement in Economic History.’’ Quarterly Journal of Economics 53:167–193. ———. 1940. ‘‘Harald Hja¨rne och den moderna historievetenskapen i Norden.’’ Historisk Tidskrift 60:133–152. ———. 1941a. ‘‘Historieforskningens objektivitet.’’ Svensk Tidskrift 118–131. ———. 1941b. Svenskt arbete och liv: fra˚n medeltiden till nutiden. Stockholm: Albert Bonniers Fo¨rlag. (This first Swedish edition was in 1957 followed by a second Swedish edition that took into account the revisions offered in the English edition of Heckscher 1954.) ———. 1947. ‘‘Ekonomisk historia och dess gra¨nsvetenskaper.’’ Historisk Tidskrift 67:1–17. ———. 1948. ‘‘Objektivititet och subjektivitet i historieforskningen.’’ In Historia och religion G. Landberg and K. G. Hildebrand, eds., Historia och religion. Stockholm: Fo¨rlag. ———. 1949a. Sveriges ekonomiska historia fra˚n Gustav Vasa. Det moderna Sveriges grundla¨ggning Tredje boken 1720–1815, Vol. 1–2. Stockholm: Albert Bonniers Fo¨rlag. ———. 1949b. ‘‘The Effect of Foreign Trade and the Distribution of Income.’’ In H. S. Ellis and L. A. Metzler, eds., Readings in the Theory of International Trade, 272–300. Philadelphia: Blakiston Company. ———. 1951a. Studium och undervisning i ekonomisk historia. Lund: C.W.K. Gleerups Fo¨rlag. ———. 1951b. ‘‘Om historiska misstag och deras behandling.’’ Historisk Tidskrift 71:1–13. ———. 1952. ‘‘David Davidson.’’ International Economic Papers 2:111–135. ———. 1953. ‘‘A Survey of Economic Thought in Sweden 1875–1950.’’ Scandinavian Economic History Review 1:105–125. ———. 1954. An Economic History of Sweden. Trans. Go¨ran Ohlin. Cambridge, MA: Harvard University Press. Henriksson, R. 1979. ‘‘Eli F. Heckscher och svensk nationalekonomi.’’ Ekonomisk Debatt 7:510–520. ———. 1987a. Montgomery, Gustaf Arthur.’’ Svenskt Biografiskt Lexikon. Stockholm: Norstedts. ———. 1987b. ‘‘Konjunkturbevakning fo¨re Konjunkturinstitutet.’’ Mimeo. Konjunkturinstitutet. ———. 1989. ‘‘The Institutional Base of the Stockholm School: The Political Economy Club 1917–1951.’’ History of Economics Society Bulletin 11:59–97. ———. 1990. ‘‘Eli Heckscher.’’ In C. Jonung and A.-C. Sta˚hlberg, eds., Ekonomportra¨tt: Svenska ekonomer under 300 a˚r, 165–186. Stockholm: SNS. ———. 1991a. ‘‘Eli F Heckscher: The Economic Historian as Economist.’’ In B. Sandelin, ed., The History of Swedish Economic Thought, 141–167. London: Routledge.
The Making of the Economic Historian: Eli F. Heckscher, 1897–1904
87
———. 1991b. ‘‘The Facts on Wicksell on the Facts: Wicksell and Economic History.’’ In J. Mokyr, ed., The Vital One: Essays in Honor of Jonathan R.T. Hughes, 33–50. Greenwich, CT: JAI Press. ———. 2001. ‘‘Nationalekonomiska Fo¨reningen.’’ Royal Economic Society Newslettter 113:11–15. ———. 2002a. ‘‘Nationalekonomiska Fo¨reningen 1877–2002.’’ Ekonomisk Debatt 30:569– 599. ———. 2002b. ‘‘Eureka Unter den Linden: A Reinterpretation of Ohlin’s Early Contributions to the Heckscher-Ohlin Theme.’’ In R. Findlay, L. Jonung, and M. Lundahl, eds., Bertil Ohlin: A Centennial Celebration, 1899–1999, 125–137. Cambridge, MA: MIT Press. ———. 2003. ‘‘Eli F Heckscher.’’ Oxford University Press Encyclopedia of Economic History. New York: Oxford University Press. Henriksson, R. G. H., and M. Lundahl. 2001. ‘‘Eli Heckscher, Economic History and Economic Theory.’’ Paper presented to the conference Trade, Development and History. Columbia University, New York, April 20–21. ———, eds. 2003. Janusansiktet Eli Heckscher: Nationalekonom och ekonomisk historiker. Kristianstad: Timbro. Hettne, B. 1980. ‘‘Ekonomisk historia i Sverige under femtio a˚r.’’ Historisk Tidskrift 100:140–175. Hildebrand, K. 1897a. ‘‘Skiljedom och allma¨n fred.’’ Meddelanden fra˚n Uppsala Fo¨rsvarsfo¨rbund 11. ———. 1897b. ‘‘Nationalekonomisk historieskrivning.’’ Historisk Tidskrift 17:171–208. Hildebrand, K.-G. 1954. ‘‘Planhusha˚llning.’’ Svenska Dagbladet, February 15, 1954. ———. 1980. ‘‘Emil Hildebrand och Historisk Tidskrift.’’ Historisk Tidskrift 100:62–91. Lane, F. C., and J. C. Riemersma, eds. 1953. Enterprise and Secular Change: Readings in Economic History. Homewood, IL: Irwin. Lundberg, E. 1952. Obituary of E. H. Heckscher. Dagens Nyheter, December 30, 1952. Magnusson, L. 1994. ‘‘Eli Heckscher and Mercantilism: An Introduction.’’ Uppsala Papers in Economic History Research Report 35. Marshall, A. 1890. Principles of Economics. London: Macmillan. Montgomery, A. 1956. ‘‘Eli F. Heckscher.’’ In J. F. Lambie, ed., Architects and Craftsmen in History: Festschrift fu¨r Abbot Payson Usher, 119–156. Tu¨bingen: Mohr (Siebeck). Nicholson, J. S. 1893–1901. Principles of Political Economy. 3 vols. London: Adam and Charles Black. Ode´n, B. 1975. Lauritz Weibull och forskarsamha¨llet. Lund: Gleerup. Olsson, C.-A. 1992. ‘‘Eli Heckscher and the Problem of Synthesis: A Methodological Note.’’ Scandinavian Economic History Review 40, no. 3: 29–52. Schmoller, G. 1919. Grundriss der allgemeinen Volkswirtshaftslehre. 2 vols. Munich: Duncker & Humblot.
88
Theory, Methodology, and History
So¨derlund, E. 1946. ‘‘Heckscher Eli Filip.’’ Svenska Ma¨n och Kvinnor 3:336–338. ———. 1953. ‘‘Eli F. Heckscher.’’ Scandinavian Economic History Review 1:137–140. Stjernberg, N. 1902. Till fra˚gan om de s k rent ekonomiska kategorierna. Uppsala: Akademiska Boktryckeriet. Uhr, C. 1979. ‘‘Eli F. Heckscher 1879–1952 and His Treatise on Mercantilism Revisited.’’ Economy and History 23:3–39. Utterstro¨m, G. 1982. ‘‘Eli Heckscher, Bertil Boethius och Sveriges Ekonomiska Historia fra˚n Gustav Vasa.’’ Meddelanden fra˚n institutionen fo¨r ekonomisk historia 2, Umea˚ Universitet. Wicksell, K. 1899. Review of Gide 1899. Ekonomisk Tidskrift 1:534–537. ———. 1901. Fo¨rela¨sningar i nationalekonomi Ha¨fte 1. Lund: Gleerup. ———. 1903. ‘‘Ma˚l och medel i nationalekonomin.’’ Ekonomisk Tidskrift 4:457–474. ———. 1907. Review of Heckscher, E. F. (1907). Statsvetenskaplig Tidskrift 10:337–341.
Part II
Heckscher-Ohlin Trade Theory
4
Eli Heckscher and the Holy Trinity Ronald W. Jones
Eli Heckscher died in 1952, just one year earlier than Wassily Leontief’s famous article appeared proclaiming what has become known as the Leontief paradox concerning the trade pattern in the United States. Received doctrine from the Swedish pair of Eli Heckscher and Bertil Ohlin had prepared us to expect that American exports would be capital-intensive in their production techniques compared with its importcompeting sector. Not so, argued Leontief, and this bombshell spurred a raft of theoretical dissertations among young students on both sides of the Atlantic as well as a plethora of empirical work that continues to this day. Fortunately for English speakers, a few years before Heckscher’s death, Svend Laursen and his wife prepared a translation from the Swedish of Heckscher’s 1919 article for the 1949 Readings in the Theory of International Trade (edited by Howard Ellis and Lloyd Metzler). This article received a revised translation in a gem of a book, Heckscher-Ohlin Trade Theory (1991), prepared and edited by Harry Flam and M. June Flanders. Although the primary objective of this latter work was to provide a translation, for the first time, of Bertil Ohlin’s 1924 dissertation, The Theory of Trade, it also served to confirm the importance of Heckscher’s pioneering article, written five years previously. It is perhaps ironic that, almost the entire academic career of Eli Heckscher was concerned with issues of economic history, whereas this single article on international trade theory was sufficient to award him the lead position in a ‘‘hyphen trophy’’ of the label, ‘‘Heckscher-Ohlin Trade Theory.’’ In the prefatory remarks to the 1949 translation of his article, Heckscher, in referring to the work of his former student, Ohlin, remarks: ‘‘My previous treatment . . . does not, in the eyes of its author, contain much of value over and above Ohlin’s books’’ (Flam and Flanders 1991:43). An innocent reader of such a remark might conclude
92
Heckscher-Ohlin Trade Theory
that Ohlin’s work appeared first, and that Heckscher was admitting that he did not have much more of substance to add. Such modesty is seen clearly to have been misplaced on rereading the Heckscher article, as well as the foreword (by Paul Samuelson) and the introduction (by Flam and Flanders) in the 1991 book. Some years ago Wilfred Ethier (1974) arranged the main substantive contributions of Heckscher-Ohlin theory into four subcategories. Two of these, the Heckscher-Ohlin theorem about the nature of trade, arguing that with technologies assumed the same everywhere trade patterns reflected differences in relative factor endowments, and the factor-price equalization theorem, positing that free trade in commodities could bring about an absolute equalization of wage rates and other factor returns even without any international mobility of the factors of production, were spelled out rather clearly in Heckscher’s original treatment. Furthermore, his discussion of the effect of tariff protection on the distribution of income foreshadows the later classical account provided by Stolper and Samuelson (1941), while his remarks on the international mobility of factors stimulated by a protective tariff can be said to anticipate some of Mundell (1957). Whereas Ohlin would later prove reluctant to accept that trade could bring about full factor-price equalization, Heckscher was perhaps too eager to push this possibility since in talking about quality variations in factors he states that ‘‘the number of factors of production is thus practically unlimited’’ (Flam and Flanders 1991:48). The problem with this is the ‘‘numbers game’’; if the number of factors exceeds the number of produced commodities, commodity prices by themselves do not determine factor returns. Factor endowments have a role to play so that trade in commodities cannot be expected to equalize the returns to factors among the trading nations. (More on this below.) Putting quality differences aside, both Heckscher and Ohlin make frequent reference to what I refer to as a classical holy trinity of factors, namely land, capital, and labor. As Samuelson remarks about Ohlin in his foreword to the Flam and Flanders volume: ‘‘Already in 1924 Ohlin has melded Heckscher and Walras. But neither then, nor in 1933 and 1967, did Ohlin descend from full generality to strong and manageable cases—such as two factors of production and two-or-more goods. What a pity. Not only did Ohlin leave to my generation these easy pickings, but in addition he would for the first time have really understood his own system had he played with graphable versions’’ (Flam and Flanders 1991:ix).
Eli Heckscher and the Holy Trinity
93
The same is not quite true of Heckscher, in that he briefly discusses a 2 2 example in which wheat and textiles are produced by labor and land. But his example is rather constrained. He assumes that with trade the price of textiles is cut in half relative to the price of wheat, and that this causes the wage rate also to be cut in half. No basis for this relationship is given. It implies that textiles only use labor (and wheat only land) and in this sense is a quite special 2 2 example.1 In the rest of these remarks I concentrate on what can be said about Heckscher-Ohlin trade theory if the classical trinity of land, capital, and labor is maintained. In short, if Heckscher had taken the route of laying out the propositions that would be valid in this three-factor case, what remarks would be justified? And how did he connect the three-factor case to the factor-price equalization possibility? The Holy Trinity with One Commodity in Each Country Most simple models of trade in the Heckscher-Ohlin tradition assume that each country can produce a (same) pair of commodities. But this is not necessary. In a two-commodity world, trade could take place with each country specialized to a different commodity. In such a case, whether there are two factors of production or three makes little difference to the conclusion that free trade will generally not result in factor price equalization. And an explanation of the trade pattern is trivial— each country exporting the sole commodity produced there, regardless of any comparison of autarky factor prices. So what extra is added by having a third factor? The existence of a third factor of production introduces the possibility that two of these factors have a complementarity relationship with each other, whereby an increase in one factor’s return would, at given output, cause the other factor to be used less intensively. (Of course, the third factor must be used more intensively in order to maintain output since less must be utilized of the factor that has gone up in price.) Even without complementarity, a factor may have a different degree of substitutability with each of the other two. As will be seen, this can matter for the issue of the effect of trade on the distribution of income, an issue of primary concern to Heckscher. The Specific Factors Model The specific-factors model, with an early treatment by Gottfried Haberler (1936) and formally developed by Jones (1971) and Samuelson
94
Heckscher-Ohlin Trade Theory
(1971), is perhaps the most simple way of sticking with the trinity of land, labor, and capital, but avoiding the complications that enter when any productive activity uses all three factors. That is, simplicity of the type familiar from 2 2 Heckscher-Ohlin models is maintained by positing that each sector uses a productive factor not used in the other while both make use of a factor that is freely mobile between sectors. Many of the subsequent uses of the model consider the pair of specific factors to be of the same general type (e.g., capital), so that over time one kind of specific factor can be transformed into the other. This interpretation suggests the specific-factors model is a short-run version of the 2 2 Heckscher-Ohlin model (e.g., as proposed by Neary 1978). Alternatively, this model might consider the three factors as distinct, as in the holy trinity. Thus suppose labor is completely mobile between sectors, capital is used only to produce manufactures, and land is used only in the production of agricultural goods. With this interpretation the model can yield strong results on the two issues of primary concern to Heckscher, namely, the effect of countries having different factor endowments on the pattern of trade and the consequences of such trade on the distribution of income. Certainly a country that is relatively land-abundant is apt to be an exporter of agricultural products, and one that is relatively capital-abundant will tend to export manufactures. As an aside, it is useful to recall that in the development of trade theory, two distinct meanings of relative factor endowments have been highlighted. On one hand is the comparison of relative physical volumes in endowments. On the other is a comparison of how relative factor returns would compare before international trade takes place. Heckscher’s interpretation of relative factor endowments clearly belongs to this latter category, and developments in Heckscher-Ohlin theory have pointed out not only that the two interpretations are different, but also that the price definition of relative factor intensity is the stronger of the two: asymmetric demand conditions in two countries could account for a physically relatively capital abundant country finding capital relatively dear in autarky if tastes there are biased heavily toward the capital-intensive good. In such a case, the pattern of trade predicted by the Heckscher-Ohlin theorem would be violated if a comparison of physical proportions was the criterion of relative factor abundance, but nonetheless would be upheld with the price version that was used by Heckscher ( Jones 1956).
Eli Heckscher and the Holy Trinity
95
Although dissimilarities either in land rents or returns to capital in autarky would have strong effects on relative commodity prices (and hence trade patterns), what is the significance of one country having relatively cheap labor before trade? As the theory points out, such a difference in the (price version of) endowment patterns has less of an effect on commodity prices because labor is used in both industries. However, by itself a lower wage rate in one country would tend to make relatively inexpensive the commodity in which labor’s distributive factor share is higher. Factor endowments all count in influencing the trade pattern, just as they do in the simple 2 2 Heckscher-Ohlin model. As suggested earlier, factor-price equalization with trade is not to be expected in this setting since the number of factors (3) exceeds the number of produced commodities (2). What can be said that is similar to results of the Stolper-Samuelson type for the 2 2 model is that returns to the specific factors are magnified reflections of any change in relative commodity prices. An increase in the relative price of manufactures, say, unambiguously improves the real return to capital and reduces that to land. The fate of laborers is subject to what has been called ‘‘the neoclassical ambiguity,’’ the nominal wage rising in terms of agriculture and falling in terms of manufactures. The magnification effects for the specific factors are reflections of the asymmetry in production technology whereby in each sector it takes both factors to produce, separately, each commodity (the assumption of no joint production). The General 3 D 2 Model Letting all three factors in the holy trinity be involved in both commodities produced is a natural extension of the specific-factors model, and was examined in detail by Raveendra Batra and Francisco Casas (1976), Katsuhiko Suzuki (1983), and Ronald Jones and Stephen Easton (1983). Just as in the earlier case of a single commodity produced, different degrees of substitutability among factors and the possibility of complementarity can no longer be avoided. Changes in the terms of trade affect the returns to each of the three factors, but the possibilities are wider than in the specific-factors model. The difference in the ranking of factor intensities by industry is, of course, less extreme than in the specific factors model. Suppose, nonetheless, that in such a ranking land is the most intensively used input in agriculture, capital in manufacturing, and once again labor is the ‘‘middle’’ factor. The effect
96
Heckscher-Ohlin Trade Theory
of commodity price changes on income distribution could be similar to that in the easier-to-analyze specific-factors case. But it need not be, since asymmetry in substitutability among factors counts. As an extreme example, suppose that land and capital are better substitutes for each other than is either with labor. This means that it would be fairly difficult for a commodity price change to alter the ratio of land’s return to that of capital by very much. But if the relative price of manufactures, say, goes up, in a competitive market place in which commodity prices are matched by unit costs, the relative cost of manufactures must rise. If labor’s distributive share in manufactures is larger than it is in agriculture, one way of having relative costs in manufactures go up to match the price increase is to have a large relative increase in the nominal wage rate, with land’s return rising very little, or perhaps falling, so as not to vary much from the return to capital. Heckscher’s Example of Factor-Price Equalization If two economies share the same technological knowledge, with each input possessing the same skills from country to country, but with different relative endowments, trade in two commodities would generally not serve to equalize factor returns because the number of commodities is smaller than the number of factors. Heckscher (Flam and Flanders 1991:54–55) discusses an example in which he seems to disagree. His discussion of this example is, in my view, rather opaque, but he seems to be saying that with common fixed coefficients in the two countries, with two commodities (meat is commodity 1, machinery is commodity 2), and with one unit of each factor required per unit of commodity 1 as the fixed input-output coefficients and, for the second commodity, one unit of land, 3 units of capital and 4 units of labor required as inputs per unit output,2 he concludes: ‘‘Exchange will thus continue until equalization is complete.’’ There are two pitfalls I see in this example. First, with rigid input/ output coefficients and only two commodities produced, there is only a restricted set of possible factor supplies that could result in full employment and thus positive factor returns. The restrictions that are required can be revealed, first, by considering the full-employment conditions for capital and labor: X1 þ 3X2 ¼ K X1 þ 4X2 ¼ L
Eli Heckscher and the Holy Trinity
97
Figure 4.1 Allowable endowments
where the Xi denote the two outputs and fixed factor endowments of capital and labor are denoted by K and L. The solutions obtained are that X1 equals ð4K 3LÞ and that X2 equals ðL KÞ. In order that these market-clearing relationships can be obtained with positive outputs of the two commodities, K must exceed 34 L but fall short of L. In addition, the market for land ðTÞ clears if X1 þ X2 ¼ T, requiring that K ¼ ð1=3ÞT þ ð2=3ÞL; and this, with the previous restriction, implies that T lies between and L. Figure 4.1 illustrates for given arbitrary labor endowment the only possible (linear) combinations of capital and land that allow full employment and positive outputs for the prescribed rigid technological coefficients. Note that it is not a unique endowment bundle; the two economies under consideration could have different endowments but identical techniques, full employment, and positive levels of output. The second problem emerges from a consideration of the competitive profit conditions of equilibrium in which costs are equal to commodity prices. Suppose the return to land, rT , is arbitrarily chosen. Then for positive equilibrium production of the two commodities:
1 4L
98
Heckscher-Ohlin Trade Theory
rK þ w ¼ ðp1 rT Þ 3rK þ 4w ¼ ðp2 rT Þ where the return to capital and the wage rate are denoted by rK and w, respectively. As for the commodity prices, pi , not only must they both exceed rT (to allow non-negative returns to the other two factors), but as well, 4ðp1 rT Þ > ðp2 rT Þ > 3ðp1 rT Þ > 0: The dependence of the return to capital and the wage rate on given commodity prices and the return to land satisfying the above set of inequalities is shown by the solutions: rK ¼ 4ð p1 rT Þ ðp2 rT Þ;
w ¼ ðp2 rT Þ 3ðp1 rT Þ:
Figure 4.2 illustrates, by the intersection point of the two lines, D, a possible joint solution for the return to capital and the wage rate given commodity prices and the return to land with both commodities produced. The heavy broken line is the factor-price frontier for wages and the return to capital. If the required inequalities are satisfied, can two economies that share this inflexible technology and face the same set of commodity prices have different factor endowments but end up with exactly the
Figure 4.2 Factor-price frontier
Eli Heckscher and the Holy Trinity
99
same factor returns without any international factor mobility being allowed? Yes. But must trade serve to equalize factor prices? No. For example, in one country there might be a slightly lower return to land than in the other country, shifting upwards the pair of linear loci in figure 4.2. Factor returns would thus be different in the two countries, with factor markets clearing in both and price equal to cost in both industries in each country. Heckscher compounds his difficulties by suggesting that in each country outputs adjust in the movement from autarky to free trade. This is not possible for a given set of factor endowments and rigid coefficients of production. Allowing smooth factor substitutability in both commodities would help in that factor prices could then change, and with them outputs of both commodities as techniques of production adjust. But the problem remains that the three-factor, two-commodity setting is not conducive to Heckscher’s conclusion that factor price equalization will be complete. In fairness to Heckscher, however, he does leave himself some wiggle room. He later mentions a third commodity (with unit input-output requirements for land, capital and labor of 1, 12 , and 15, respectively) (see Flam and Flanders 1991:55–56), and earlier in his two-commodity illustration he does not state explicitly that these are the only two commodities produced. It was left to Samuelson and followers to point out more explicitly the balance between the number of factors and commodities required to get factor price equalization. As well, he does link factor-price equalization to the situation in which ‘‘techniques of production are the same in all countries’’ (54). If techniques were flexible, it would be the case that factor prices are uniquely linked to the techniques chosen, so that factor returns would be lined up between countries if, with trade, these techniques were the same. However, in his example the techniques were inflexible and assumed to be the same between countries, and this inflexibility disrupts the link between techniques and factor prices. Another problem with the numbers chosen by Heckscher in his illustration is revealed by a closer analysis of figure 4.2. In his discussion of how the United States and Europe differ in factor returns (58–59), he stresses that land rents are low in the United States and thus wages are high—fostering immigration from Europe if international trade in commodities does not suffice to equalize factor prices. Would a country with the technology assumed by Heckscher experience an increase in labor’s wage rate if the return on land were to be reduced? Both loci in figure 4.2 would be shifted upward, but in such a way that the
100
Heckscher-Ohlin Trade Theory
return to capital rises and the wage rate falls. To use more modern terminology, the reduction in land rents lowers the relative effective price in the second sector—that is, ð p2 rT Þ=ðp1 rT Þ—and the second sector is relatively labor intensive. The 3 D 3 Case The objections raised above to the argument presented by Heckscher to establish the proposition that international trade suffices to bring equality of factor prices among countries even without any international factor mobility (as long as they share a common technology) cease to have relevance if the number of productive inputs is matched by the number of produced commodities, as in the 3 3 case. Even if production coefficients are rigid, alterations in the composition of commodity outputs often suffice to clear factor markets when factor endowments are altered, even without any change in factor prices. The qualification refers to the necessity (in order to clear markets) that the vectors of techniques for the three commodities contain the factor endowment vector. (In the 2 2 case, this is stated as requiring the endowment vector to lie within the cone of diversification).3 If so, two countries with different endowments would find their factor returns brought to equality with free trade. And Heckscher’s argument would be even stronger if the common technologies were flexible in allowing techniques actually adopted to be sensitive to factor prices. Even if technologies are flexible, there is no guarantee that in this 3 3 case a country will actually produce all three commodities in a free-trade equilibrium, although it could do so in the absence of international trade. Perhaps the most basic feature of trade is that it allows a country to consume many more commodities than it produces. In the present scenario a country might produce all three commodities, or perhaps just two, or, indeed, may be specialized completely to the commodity in which it has the greatest comparative advantage. Two countries that share the same technology but have different factor endowments may thus very well produce different commodities, or perhaps have an overlap of a commodity produced in common with a commodity that is not. Heckscher was well aware of this possibility, and he stressed that in such a case factor returns would not be equalized by trade, any commodity produced in common would not have the same technique utilized in each country, and that international migration would result if allowed. He discussed this possibility in the context of trade between the United States and Europe.
Eli Heckscher and the Holy Trinity
101
There is a reservation about trade and factor-price equalization that was not brought out explicitly by Heckscher, but was the focus of much of the literature some decades later. That reservation has to do with the possibility of factor-intensity reversals. This phenomenon is typically discussed in the 2 2 setting. Although a country’s factor prices are usually assumed uniquely to determine productive techniques (i.e., no ‘‘flats’’ on the isoquants) and thus the cost of production (equal to price in a competitive equilibrium in which the commodity is produced), uniqueness may not work in the other direction. That is, commodity prices may not uniquely determine factor prices. (An early discussion of this is found in Samuelson’s factor-price equalization articles [1948, 1949]). A pair of countries might each produce the full complement of commodities (that is, the same number as the number of factors), but their endowment vectors could lie in different cones of diversification. Although much of the mathematical literature concerned with the conditions for the univalence of the commodity pricefactor price relationship pays no attention to factor endowments, trade theorists from Samuelson’s 1948 article have always stressed that a necessary condition for factor-price equalization is that countries’ endowments be fairly similar. The questions most intensely debated in the literature concerning the 3 3 scenario have to do with the pair of ‘‘Heckscher-Ohlin propositions’’ not stressed by Heckscher, nor by Ohlin, namely, the StolperSamuelson theorem (1941) and the Rybczynski theorem (1955). Are there strong general results that can be stated for the effect of a commodity price change on the distribution of factor income or for the effect of endowment changes on the composition of output for a country facing given commodity prices? In a sense the answer is in the negative, in that more structural detail is required, having to do with factor intensity comparisons. For example, John Chipman (1969) showed that an increase in any commodity price would unambiguously raise the real return to the factor most intensively used in that sector if that factor’s distributive share in that sector exceeded its share in the other two industries.4 Heckscher did not explicitly concern himself with either of these propositions. The Stolper-Samuelson kind of result rests heavily upon the assumption that productive activities are not joint. That is, in each a combination of inputs yields a unique individual output. The factor-price equalization result, I would argue, does not ( Jones 1992b).5 Of course the Heckscher-Ohlin theorem, discussing the relationship between factor endowments and trade patterns, is related to the Rybczynski type of theorem. In the 3 3 case, Ed Leamer (1987 and 1994)
102
Heckscher-Ohlin Trade Theory
has extensively explored possible relationships among physical factor endowments and trade patterns as well as trade and factor returns.6 In this context it is useful to recall that Heckscher adhered to the price version of relative factor endowments. The Holy Trinity and Many Commodities I have often maintained that Heckscher-Ohlin theory can best be seen (and is more easily understood) in the context in which many commodities can technically be produced. The reason for this is that with trade a country can be highly specialized, even to the extent of producing only a single commodity. Thus the interesting question can be raised: With trade, what commodities does a country produce? This is Ricardian in its tone. If world prices are not necessarily reflected in a country’s own technology, the country need not produce more commodities than it has productive factors, and may produce an even smaller number. Thus, if there are more commodities than three, there is much room for production patterns to differ among countries and for their factor returns to be different even with commodity trade. The stage is set for a discussion of international mobility of productive factors, and Heckscher was very much involved with discussions of Swedish emigration. This is a setting in which Heckscher’s unique and great role in trade theory can well blend in with his voluminous published work on economic history. Concluding Remarks There is no doubt that Heckscher’s 1919 article is a classic. Since its translation into English (thirty years after its publication) it became clear that Eli Heckscher’s contribution to the modern theory of international trade (as it was then called) had to be put on a par with that of his brilliant student, Bertil Ohlin. Ohlin added more technical material, borrowing from Cassel (and, indirectly, from Walras), and, especially in his 1933 Harvard University Press volume, delved deeply into historical episodes as well as into the importance of economies of scale. (This was later picked up in so-called new trade theory. See, for example, Paul Krugman’s remarks [1999]). In his earlier and shorter presentation, Heckscher also brought up the possibility that scale in production was important, and could lead to lower costs and better technology.
Eli Heckscher and the Holy Trinity
103
In retrospect, perhaps the major difference in the contributions of these two Swedish economists concerned the effect of international trade on a nation’s distribution of income. Although I have argued above that Heckscher’s illustration of why factor prices might be fully equalized with trade is subject to reservations, he was absolutely clear that such a result was possible. In particular, he was explicit on the fact that having factor prices brought to equality with trade would not signal the end of trade. Instead, trade would merely have reached its equilibrium level and would not expand further (unless conditions changed). In this he was bolder than Ohlin, who, perhaps with a keen eye on actual trading conditions, was reluctant to accept that international trade by itself could bring about full equalization of factor returns. Heckscher remarked on the possible mobility of factors between countries in case countries were sufficiently dissimilar that trade would not equate factor returns. As well, he considered the case in which changes in factor returns would alter their domestic supply. In this he was correct in asserting that variable factor supplies would result in greater differences in factor endowments between countries as a consequence of trade (which would serve to lessen the possibility of factor price equalization). I was first introduced to the English translation of Heckscher’s article when Heckscher was still alive. It was tough reading then, and it remains so for me to this day. As Samuelson has remarked, both Heckscher and Ohlin missed the opportunity to spell out their model in simple terms, such as in the two-commodity, two-factor case that proved of such worth in countless succeeding articles and books. (But I am thankful for the great employment opportunities this provided to subsequent generations.) I think this reflects the nature of the times. After all, which of the holy trinity of labor, capital, and land could be denied, especially in a time and place in which each of these factors was deemed to be so important in production, especially by an economic historian? With the holy trinity maintained, Heckscher still did a remarkable job in laying out the foundations of what has since become known as Heckscher-Ohlin theory. Notes 1. The assumption that wheat uses only land and no labor is necessary for Heckscher’s remark that the wage rate is cut in half relative to the rental on land. If wheat uses labor as well as land in this 2 2 example, and the price of wheat is constant, the rental on land
104
Heckscher-Ohlin Trade Theory
would be driven up, and thus the wage rate would be cut by more than half relative to land’s rental. 2. Curiously enough, the input coefficients of capital and labor in machinery (industry 2) cited by Flam and Flanders are 4 and 5 (instead of 3 and 4). The latter pair of numbers corresponds to the 1949 translation and, according to Mats Lundahl, to the original Swedish 1919 article as well. 3. See Chipman (1966). 4. In the same issue of the journal in which Chipman’s contribution appeared, Murray Kemp and Leon Wegge (1969) provided even more strict conditions on factor shares that would suffice, in the 3 3 case, to establish that an increase in any commodity price would raise the real return to the factor used intensively there and lower the returns to the other two factors. In both of these articles the criteria used to establish results in the 3 3 case were shown not to be sufficient in the 4 4 case. Sufficient conditions for each of these cases in higher dimensions were provided later in Jones, Marjit, and Mitra (1993) and Mitra and Jones (1999). 5. For a divergent view, see Samuelson (1992). 6. For agreements and disagreements with Leamer, see Jones (1992a).
References Batra, R., and F. Casas. 1976. ‘‘A Synthesis of the Heckscher-Ohlin and the Neoclassical Models of International Trade.’’ Journal of International Economics 6:21–38. Chipman, J. S. 1966. ‘‘A Survey of the Theory of International Trade, Part 2.’’ Econometrica 33:685–760. ———. 1969. ‘‘Factor Price Equalization and the Stolper-Samuelson Theorem.’’ International Economic Review 10:399–406. Ellis, H., and L. Metzler, eds. 1949. Readings in the Theory of International Trade. London: George Allen and Unwin. Ethier, W. 1974. ‘‘Some of the Theorems of International Trade with Many Goods and Factors.’’ Journal of International Economics 4:199–206. Flam, H., and M. J. Flanders, eds. 1991. Heckscher-Ohlin Trade Theory. Cambridge, MA: MIT Press. Haberler, G. 1936. The Theory of International Trade. London: William Hodge. Heckscher, E. 1919. ‘‘The Effect of Foreign Trade on the Distribution of Income.’’ Ekonomisk Tidskrift 21:497–512. Jones, R. W. 1956. ‘‘Factor Proportions and the Heckscher-Ohlin Theorem.’’ Review of Economic Studies 24:1–10. ———. 1971. ‘‘A Three-Factor Model in Theory, Trade and History.’’ In J. Bhagwati et al., eds., Trade, Balance of Payments and Growth, 3–21. Amsterdam: North-Holland. ———. 1992a. ‘‘Factor Scarcity, Factor Abundance and Attitudes Towards Protection: The 3 3 Model.’’ Journal of International Economic Integration 7:1–19.
Eli Heckscher and the Holy Trinity
105
———. 1992b. ‘‘Jointness in Production and Factor-Price Equalization.’’ Review of International Economics 1:10–18. Jones, R. W., and S. Easton. 1983. ‘‘Factor Intensities and Factor Substitution in General Equilibrium.’’ Journal of International Economics 15:65–99. Jones, R. W., S. Marjit, and T. Mitra. 1993. ‘‘The Stolper-Samuelson Theorem: Links to Dominant Diagonals.’’ In Robert Becker et al., eds., General Equilibrium, Growth and Trade II, 429–441. San Diego: Academic Press. Kemp, M. C., and L. Wegge. 1969. ‘‘On the Relation Between Commodity Prices and Factor Rewards.’’ International Economic Review 10:407–413. Krugman, P. 1999. ‘‘Was It All in Ohlin?’’ In Ronald Findlay et al., eds., Bertil Ohlin: A Centennial Celebration, 389–405. Cambridge, MA: MIT Press. Leamer, E. 1987. ‘‘Paths of Development in the Three-Factor n-Good General Equilibrium Model.’’ Journal of Political Economy 95:961–999. ———. 1994. ‘‘Commemorating the Fiftieth Birthday of the Stolper-Samuelson Theorem.’’ In Alan Deardorff and Robert Stern, eds., The Stolper-Samuelson Theorem: A Golden Jubilee, 289–308. Ann Arbor: University of Michigan Press. Leontief, W. 1953. ‘‘Domestic Production and Foreign Trade: The American Capital Position Re-examined.’’ Proceedings of the American Philosophical Society 97:332–349. Mitra, T., and R. W. Jones. 1999. ‘‘Factor Shares and the Chipman Condition.’’ In James Melvin et al., eds., Trade, Welfare and Econometrics, 135–143. New York: Routledge. Mundell, R. A. 1957. ‘‘International Trade and Factor Mobility.’’ American Economic Review 47:321–335. Neary, J. P. 1978. ‘‘Short-run Capital Specificity and the Pure Theory of International Trade.’’ Economic Journal 88:488–510. Ohlin, B. 1924. The Theory of Trade (Handelns Teory), in Flam and Flanders 1991:75–214. ———. 1933. Interregional and International Trade. Cambridge, MA: Harvard University Press. Rybczynski, T. 1955. ‘‘Factor Endowments and Relative Commodity Prices.’’ Economica 22:336–341. Samuelson, P. A. 1948. ‘‘International Trade and the Equalisation of Factor Prices.’’ Economic Journal 58:163–184. ———. 1949. ‘‘International Factor-Price Equalisation Once Again.’’ Economic Journal 59:181–197. ———. 1971. ‘‘Ohlin was Right.’’ Swedish Journal of Economics 73:365–384. ———. 1992. ‘‘Factor-Price Equalization by Trade in Joint and Non-joint Production.’’ Review of International Economics 1:1–9. Stolper, W., and P. A. Samuelson. 1941. ‘‘Protection and Real Wages.’’ Review of Economic Studies 9:58–73. Suzuki, K. 1983. ‘‘A Synthesis of the Heckscher-Ohlin and the Neoclassical Models of International Trade: A Comment.’’ Journal of International Economics 14:141–144.
5
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization Kevin H. O’Rourke
The factor proportions theory of trade developed by Eli Heckscher and Bertil Ohlin is so intuitively appealing that it remains the bedrock of modern trade courses. Despite its popularity, however, doubts have persistently been raised about its empirical applicability, from the Leontief paradox (Leontief 1953) to the stylized facts (high levels of intra-industry trade; high levels of trade between similar countries) that motivated the development of new trade theory in the 1980s. In an influential paper that was particularly damaging to the theory’s credibility, Bowen, Leamer, and Sveikauskas (1987) showed that the Heckscher-Ohlin-Vanek model was of no help when trying to predict the net factor content of a country’s trade. However, recent work by Don Davis, David Weinstein, and others has suggested that Heckscher-Ohlin theory does indeed help to explain trade patterns, so long as the researcher bears in mind the fact that countries are not distinguished by differences in factor endowments alone; for example, Davis and Weinstein (2001) show that HeckscherOhlin-Vanek theory is consistent with the data, as long as it is modified to take account of the (self-evidently true) facts that technology differs across countries, that factor price equalization does not hold, that some goods are nontraded, and that international trade is not costless. In this essay, I take an entirely different approach in assessing the empirical usefulness of Heckscher-Ohlin theory. I do not ask whether it explains trade patterns, which is what the theory is supposed to do; rather, I ask whether individuals’ attitudes toward globalization (and more specifically, their attitudes toward trade and immigration) are consistent with factor proportions theory. In particular, I start from the premise that trade and migration patterns are today driven largely by differences in the relative endowments of skilled and unskilled labor in different countries; it follows from Heckscher-Ohlin theory that
108
Heckscher-Ohlin Trade Theory
skilled and unskilled workers should differ in their attitudes towards globalization, in a predictable manner. Do these predictions hold true when confronted with the data? In a series of papers, Kenneth Scheve and Matthew Slaughter have addressed these issues and found evidence for the Heckscher-Ohlin worldview. Scheve and Slaughter (2001a) examined individual-level survey data for the United States in 1992 and found that low-skilled workers were more likely to support ‘‘new limits on imports’’ than high-skilled workers; they also found that factor type (i.e., skill level) was more important than the sector in which individuals were employed in explaining preferences. This finding was consistent with Heckscher-Ohlin models in which factors of production are mobile between sectors, but inconsistent with specific factors models in which agents are intersectorally immobile. Scheve and Slaughter (2001b) use U.S. survey data for 1992, 1994, and 1996 to examine attitudes toward immigration. They find that high-skill workers are less likely to support restrictionist immigration policies than their low-skill counterparts. While such findings may be consistent with a Heckscher-Ohlin worldview, single country studies cannot convincingly demonstrate that factor proportions models are relevant in explaining individual preferences regarding globalization. The reason is straightforward: Heckscher-Ohlin theory predicts that the impact of skill on attitudes should vary in a systematic way across countries. In skill-abundant countries, high-skill workers should favor trade; in low-skill-abundant countries, it is the unskilled who should favor trade. The Scheve and Slaughter findings, on their own, do not preclude the possibility that the high-skilled are in favor of globalization everywhere—for example, because better educated people understand the intellectual arguments in favor of international integration. Such a world would be at dramatic variance with the predictions of Heckscher-Ohlin theory. In order to test the theory, therefore, we need data giving attitudes toward globalization in a number of different countries. The crucial issue then becomes whether the relationship between skills and attitudes varies across countries in a manner consistent with theory. This paper will survey recent attempts to do precisely this, looking separately at attitudes toward trade and attitudes towards immigration, and drawing on my work with Richard Sinnott (O’Rourke and Sinnott 2001, 2006).1 The next sections will discuss what theory has to say about how individuals in different countries should feel about trade and immigration; the data used; the determinants of individual attitudes
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
109
toward trade; the determinants of individual attitudes toward immigration; and the determinants of attitudes toward both trade and immigration. Theoretical Expectations Trade Standard Heckscher-Ohlin trade theory is quite clear in its predictions regarding who should benefit and who should lose from free trade in commodities. Imagine a two-factor world in which countries are distinguished only by their relative endowments of skilled and unskilled workers. The relative wages of skilled workers will be lower, other things being equal, in skill abundant countries (which we will denote by R and refer to as rich countries) than in unskilled labor abundant countries (denoted by P and referred to as poor countries): we have ðwS =wUS Þ R < ðwS =wUS Þ P , where wS and wUS denote skilled and unskilled wages respectively. It is this inequality that drives comparative advantage: the rich countries will export skill-intensive goods, while the poor countries will export unskilled labor-intensive goods. The result is then relative factor price convergence (or, in the limit, factor price equalization): when countries move toward freer trade, the relative price of skilled labor rises in rich countries and falls in poor countries. Moreover, the abundant factor gains in real terms in all countries, while the scarce factor loses. Thus the skilled should favor free trade in rich countries, while they should favor protection in poor countries; the unskilled in rich countries should favor protection, while the unskilled in poor countries should support free trade. Note that Heckscher-Ohlin theory argues that individuals’ interests are related to countries’ factor endowments; in order to test the theory, we ought in principle to see whether the relationship between skills and protectionist sentiment varies across countries in a manner related to their skill endowments (for example, their average educational levels). For reasons outlined later, however, the available educational data are not satisfactory, and in testing the theory we assume that GDP per capita is strongly and positively correlated with human capital endowments. We therefore have Prediction 1: that the impact of skills on protectionist sentiment should be related to a country’s GDP per capita. In the richest countries, being high-skilled should have a negative impact on protectionist sentiment. In the poorest countries,
110
Heckscher-Ohlin Trade Theory
being high-skilled should have a positive impact on protectionist sentiment. More generally, an interaction term between skills and GDP per capita should enter with a negative sign in a regression explaining protectionist sentiment. Immigration In a pure Heckscher-Ohlin world in which technology is identical across countries, and in which countries are only distinguished by their relative endowments of skilled and unskilled labor, it is again possible to make unambiguous predictions about who should favor immigration and who should not. This is the case, even though international migration is not driven by comparative advantage and relative factor prices, but by absolute advantage, and by absolute factor price differentials. In a pure Heckscher-Ohlin world, the real wages of skilled workers will be higher in poor countries (where skilled workers are scarce) than in rich countries (where they are abundant), while unskilled wages will be higher in rich countries than in poor countries: R P we have (in real terms) wSP > wSR , but wUS > wUS . Thus, we should observe skilled workers migrating from rich to poor countries, and unskilled workers migrating from poor to rich countries. Immigration will hurt skilled workers in poor countries, but benefit the unskilled there; therefore in poor countries the unskilled should favor immigration, while skilled workers should oppose it. The situation is the reverse in rich countries: immigration will hurt the unskilled, but benefit skilled workers. Thus skilled workers should be pro-immigration, while the unskilled should oppose it. We thus have Prediction 2: the impact of skills on anti-immigrant sentiment should be related to a country’s GDP per capita. In the richest countries, being high-skilled should have a negative impact on anti-immigrant sentiment. In the poorest countries, being high-skilled should have a positive impact on anti-immigrant sentiment. More generally, an interaction term between skills and GDP per capita should enter with a negative sign in a regression explaining anti-immigrant sentiment. Note that in such a pure two-country, two-factor Heckscher-Ohlin world, in which countries are distinguished solely by their relative factor endowments, agents are consistent in their attitudes toward globalization. That is, in rich countries skilled workers favor both trade
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
111
and immigration, while unskilled workers are protectionist and antiimmigration. In poor countries, it is the unskilled who are liberal in their attitudes toward both trade and immigration, while the skilled favor both protection and immigration restrictions. This symmetry reflects the fact that in a pure two-factor Heckscher-Ohlin world in which technology is identical across countries, trade and factor flows are substitutes: they have identical effects on factor prices (i.e., they both lead to relative and absolute factor price convergence), and thus, the more you have of one dimension of globalization, the less incentive there will be for the other dimension to take place. In such a world, scarce factors lose as a result of either trade or immigration, while abundant factors gain from either. One immediate political consequence of the fact that trade and migration are substitutes for each other is that agents who are protectionist should also be anti-immigration: both trade and immigration have to be simultaneously restricted, since either phenomenon will hurt the scarce factor. Protection without immigration restrictions will not work, since protection without immigration restrictions will simply lead to more immigration; immigration barriers without protection will not work, since immigration barriers on their own will simply lead to more trade (Mundell 1957). We thus have Prediction 3: all other things being equal, being protectionist should increase the likelihood that an individual is antiimmigrant; while being anti-immigrant should increase the likelihood that an individual is protectionist. Things get a lot more complicated if technology differs across countries, or if there are more than two factors of production. Any test of Heckscher-Ohlin theory will in all likelihood do better by admitting such possibilities (recall that it is precisely by admitting the existence of such complications that the empirical trade literature has to some extent rehabilitated the theory in recent years). If technology is superior in the rich country, or if the rich country is better endowed with some third factor of production than the poor country, then it no longer follows from an inequality such as ðwS =wUS Þ R < ðwS =wUS Þ P that skilled workers will migrate from rich to poor countries: it is quite possible that ðwS =wUS Þ R < ðwS =wUS Þ P , but that (in real terms) wSR > wSP . In this case, skilled workers will move from poor (unskilled labor abundant) countries to rich (skill abundant) countries: unskilled workers will move in the same direction as skilled workers. This is, of course, what happens in the real world, suggesting that richer countries do
112
Heckscher-Ohlin Trade Theory
indeed enjoy superior technology to poor countries, and that endowments alone cannot explain differences in income, or for that matter trade patterns and factor flows. The issue of whether skilled or unskilled workers should be more anti-immigration in rich countries thus becomes unclear. Presumably it depends upon whether immigration predominantly involves skilled or unskilled workers; but which is true is not immediately obvious.2 In fact, there is a large theoretical literature which asks whether migrants are more likely to be skilled or unskilled, but this literature tends not to be located within standard Heckscher-Ohlin trade models. For example, Katz and Stark (1984) argue that asymmetric information can lead to migration flows disproportionately involving unskilled workers, since employers in rich countries may not be able to correctly discern the skill levels of potential migrants, although the equilibrium outcome can change if various devices reinstating informational symmetry are employed (Katz and Stark 1987). While appealing, it is not clear to me how this theory could be empirically tested with the data at my disposal. An alternative theory is provided by Borjas (1987), who adapts Roy’s (1951) model of occupational self-selection to the issue of migration. The conclusion of the analysis is that there will be positive selfselection of migrants if (a) the correlation between the earnings which they receive in the home and destination countries is sufficiently high; and (b) if income is more dispersed in the destination country than in the home country. On the other hand, there will be negative selfselection if (a) the correlation between the earnings which they receive in the home and destination countries is sufficiently high; and (b) if income is less dispersed in the destination country than in the home country. The theory thus predicts that immigrants into more unequal countries should be higher-skilled than immigrants into more equal countries: it follows that the high-skilled should be less favorably disposed toward immigrants in more unequal countries than in more equal countries. We have Prediction 4: the impact of skills on antiimmigrant sentiment should be related to a country’s level of inequality. In the most unequal countries, being high-skilled should have a positive impact on anti-immigrant sentiment. In the most equal countries, being high-skilled should have a negative impact on antiimmigrant sentiment. More generally, an interaction term between skills and inequality should enter with a positive sign in a regression explaining anti-immigrant sentiment.3
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
113
The Data The 1995 International Social Survey Programme (ISSP) module on national identity provides the kind of cross-country survey data4 that are needed to test the hypotheses outlined above. The ISSP national identity survey was conducted in twenty-four countries in 1995–96. The countries concerned were: Australia, West Germany and East Germany (divided again for purposes of the survey), Great Britain, the United States, Austria, Hungary, Italy, Ireland, the Netherlands, Norway, Sweden, the Czech Republic, Slovenia, Poland, Bulgaria, Russia, New Zealand, Canada, the Philippines, Japan, Spain, Latvia, and Slovakia. The survey provides two questions that are relevant in assessing attitudes toward globalization. The first asks respondents how much they agree or disagree with the statement that their country ‘‘should limit the import of foreign products in order to protect its national economy.’’ The second asks respondents if the number of immigrants to their economy should be increased a lot (1), be increased a little (2), remain the same (3), be reduced a little (4), or be reduced a lot (5). Table 5.1 reports the mean response to these questions in each country: a score greater than 3 indicates that on average respondents were leaning towards greater restriction, rather than freer trade or immigration. In every country in the sample, respondents on average favored lowering the number of immigrants; in every country in the sample bar two (the Netherlands and Japan) respondents on average favored limiting imports. Answers to these two questions constitute the dependent variables that are to be explained in the analysis which follows. The data set also provides individual-level measures of a range of demographic, socioeconomic, and political variables that are of relevance in understanding attitudes toward globalization. Among the socioeconomic variables, the most valuable from the point of view of testing the implications of the theories surveyed earlier is the respondent’s skill level. This is arrived at by coding the answers to questions on respondents’ occupation using the International Labor Organization’s ISCO88 (International Standard Classification of Occupations) coding scheme. While a complex coding scheme of this sort allows for very fine distinctions between different occupations, it makes most sense to focus on the four main skill categories provided by ISCO88. In brief, these are: (1) ‘‘elementary occupations’’ (i.e., ‘‘manual labor and simple and
114
Heckscher-Ohlin Trade Theory
Table 5.1 Summary statistics, selected variables Protect
Anti-immigrant
Country
Mean
Std. Dev.
Mean
Std. Dev.
Australia
3.997
0.988
3.768
1.042
W. Germany
3.083
1.232
4.226
0.910
E. Germany
3.563
1.189
4.338
0.871
Britain
3.723
1.004
4.052
0.962
USA
3.707
1.016
3.873
1.044
Austria
3.873
1.163
3.804
0.933
Hungary
4.047
1.075
4.402
0.817
Italy Ireland
3.571 3.65
1.216 1.128
4.151 3.071
0.900 0.829
Netherlands
2.912
0.992
3.826
0.924
Norway
3.144
1.038
3.847
0.982
Sweden
3.228
1.081
3.961
1.017
Czech Rep.
3.415
1.294
4.158
0.880
Slovenia
3.465
1.174
3.939
0.868
Poland
3.787
1.083
3.888
1.060
Bulgaria Russia
4.190 3.670
1.09 1.282
4.219 3.717
0.990 0.971
New Zealand
3.406
1.147
3.742
1.053
Canada
3.264
1.135
3.317
1.135
Philippines
3.624
0.918
3.796
1.102
Japan
2.919
1.282
3.391
1.008
Spain
3.813
0.906
3.401
0.813
Latvia
4.042
1.18
4.182
0.884
Slovakia
3.488
1.273
4.004
0.911
Source: Data from ISSP National Identity Survey (1995).
routine tasks, involving . . . with few exceptions, only limited personal initiative’’ [ILO 1990:7]); (2) ‘‘plant and machine operators and assemblers; craft and related trades workers; skilled agricultural and fishery workers; service workers and shop and market sales workers; clerks’’; (3) ‘‘technicians and associate professionals’’; and (4) ‘‘professionals.’’ A fifth group, ‘‘legislators, senior officials, and managers,’’ do not have a skill coding under this four-step skill classification and were included as a separate, fifth, skill category. Finally, members of the armed forces were excluded, since it was unclear what their skill levels were. Skill data were available for twenty-one of the twenty-four countries; the
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
115
other three (Italy, Sweden, and Japan) were omitted when estimating models involving skill. The analysis also uses a subjective economic variable, namely the stated willingness of people to move from one location to another in order to improve their standard of living or their work environment. Respondents were asked: ‘‘If you could improve your work or living conditions, how willing or unwilling would you be to move to another neighbourhood or village; another town or city within this county or region; another county or region; outside [named country]; outside [named continent]?’’ Based on the responses to these questions, two binary variables were derived, indicating whether or not individuals were nationally mobile, and internationally mobile.5 Arguably, those willing to relocate within the country should be less affected by any dislocation implied by immigration or free trade than those who are immobile. This will be particularly true if national labor markets are not perfectly integrated; and if immigrants tend to concentrate in particular regions or cities, or if import-competing industries are similarly concentrated. The rationale behind including the international mobility variable is to test Rodrik’s (1997) argument that globalization is currently favoring internationally mobile factors of production (i.e., physical and human capital) over immobile factors such as unskilled labor; alternatively (in the context of migration), being willing to live overseas may signal an openness to other cultures, and hence a greater tolerance for immigrants. The survey also indicates whether the respondent had ever lived abroad; previous experience of living abroad may provide a signal regarding willingness to move again (or, again in the context of immigration, it may indicate familiarity with foreigners). In addition, the survey provides information on respondents’ age; their gender; their religion; on whether they and their parents are native-born or not; on their marital and employment status; and on a variety of other personal characteristics and attitudes. The ISSP national identity data set includes a wide range of indicators of nationalist attitudes. The analysis here focuses on the following seven questions (versions implemented in Ireland, other country/ nationality labels substituted as appropriate): ‘‘Generally speaking, Ireland is a better country than most other countries’’
‘‘The world would be a better place if people from other countries were more like the Irish’’
116
Heckscher-Ohlin Trade Theory
‘‘I would rather be a citizen of Ireland than of any other country in the world’’
‘‘It is impossible for people who do not share Irish customs and traditions to become fully Irish’’
‘‘People should support their country even if the country is in the wrong’’
‘‘Ireland should follow its own interests, even if this leads to conflicts with other nations’’
‘‘How important do you think each of the following is for being truly Irish?’’ . . . ‘‘to have been born in Ireland’’
In each case, respondents were asked to rank their responses along a scale, in the case of the first six items, from 1 (strongly disagree) to 5 (strongly agree) and, in the case of the seventh item, from 1 (very important) to 4 (not at all important). The seventh item was reordered to make it consistent with the other six. Principal components analysis of these responses yielded two factors or underlying dimensions of nationalist attitudes. As can be seen from the rotated factor loadings in table 5.2, the first factor is a straightforward preference for and sense of the superiority of one’s own country (here labeled patriotism). The second factor identifies a narrow or exclusive sense of nationality comTable 5.2 Factor analysis of nationalist items in ISSP National Identity Survey (1995) Factor 1
Factor 2
[COUNTRY] better country than most other countries
0.86
0.02
World better place if people from other countries more like the [NATIONALITY]
0.78
0.2
Rather be citizen of [COUNTRY] than of any other country in world
0.61
0.29
Impossible for people who do not share [NATNL.] traditions to be fully [NATNL.]
0.01
0.71
People should support their country even if country is wrong
0.20
0.63
Importance of having been born in [COUNTRY] to be fully [NATIONALITY]
0.16
0.63
[COUNTRY] should follow own interests, even if conflicts with other nations
0.23
0.55
26.34
24.50
Percent variance
Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. Source: O’Rourke and Sinnott (2001). Data from ISSP National Identity Survey (1995).
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
117
bined with a degree of chauvinism of the ‘‘my country right or wrong’’ variety (here labeled chauvinism). On the basis of this analysis, patriotism and chauvinism scores have been calculated by averaging responses across the relevant subsets of items identified in the factor analysis. Finally, in order to test the various hypotheses outlined in the previous section, data on GDP per capita in 1995 were collected from the World Bank’s World Development Indicators (these are PPP-adjusted figures, in 1995 international dollars); the same source yielded information on inequality (namely, Gini coefficients).6 Data on educational attainments are also available; the standard source is the Barro and Lee (2000) dataset on average years of schooling in each country. However, the transition countries account for nine of the twenty-four countries in the ISSP dataset, and the Barro-Lee figures for schooling in several transition countries are very high: for example, average schooling is higher in Slovakia, Bulgaria, Latvia, and Poland than in the Netherlands, Ireland, and Austria. It seems doubtful that these figures provide a genuine reflection of the economically relevant human capital endowments of these countries; it is for this reason that GDP per capita data are used when testing Heckscher-Ohlin theory. Understanding Protectionist Preferences Table 5.3 presents results of a series of ordered probit regressions in which the dependent variable is ‘‘protect,’’ an ordered variable running from 1 (least protectionist) to 5 (most protectionist). The results differ from those presented in O’Rourke and Sinnott (2001) in that the specification of the equations is altered to make them more comparable with the results for anti-immigrant sentiment.7 Equation (1) shows that nationalist sentiment is an extremely strong determinant of attitudes toward trade, with patriotism, and especially chauvinism, having a large positive effect on protectionist sentiment. This result is robust across all specifications, and confirms the importance of ideology in determining attitudes towards globalization. Is there also a role for interests in shaping voter preferences? The other equations suggest that there is. Equation (2) provides a test of Prediction 1. It adds a skill variable, Skill345, to the specification, as well as an interaction term between Skill345 and GDP per capita. Skill345 is a variable indicating whether the respondent is high-skilled or not; it is equal to one if the respondent belongs to one of the three
Table 5.3 Determinants of protectionist preferences (ordered probit) (dependent variable: protect) (1)
(2)
(3)
(4)
Patriotism
0.2009*** [0.0182]
0.1954*** [0.0197]
0.1767*** [0.0173]
0.1605*** [0.0159]
Chauvinism
0.3559*** [0.0203]
0.3483*** [0.0213]
0.3385*** [0.0219]
0.3044*** [0.0198]
Skill345
0.0378 [0.0802]
0.0287 [0.0777]
0.0518 [0.0858]
Skill345*GDPCAP
0.0088** [0.0042]
0.0089** [0.0039]
0.0069 [0.0044]
National mobility
0.0066 [0.0234]
0.0164 [0.0240]
International mobility
0.0809*** [0.0227]
0.0728*** [0.0270]
Never lived abroad
0.0863*** [0.0201]
0.0855*** [0.0200]
Native
0.0027 [0.0561]
0.0276 [0.0640]
Native parents
0.0269 [0.0600]
0.0055 [0.0584]
Age
0.0122*** [0.0037]
0.0105*** [0.0039]
0.0001*** [0.0000]
Age squared
0.0001** [0.0000]
Female
0.1854*** [0.0296]
0.1906*** [0.0283]
Married
0.0178 [0.0157]
0.0046 [0.0196]
Catholic
0.0587*** [0.0177]
0.0636*** [0.0170]
Unemployed
0.0676** [0.0291]
0.0834*** [0.0278]
Anti-immigrant Cut1
0.1325*** [0.0136] 0.1478* [0.0798]
0.2508*** [0.0951]
0.1814 [0.1584]
0.3936** [0.1621]
Cut2
0.7823*** [0.0646]
0.7348*** [0.0570]
1.1821*** [0.1161]
1.4377*** [0.1258]
Cut3
1.4824*** [0.0659] 2.4691*** [0.0887]
1.4147*** [0.0573] 2.4185*** [0.0801]
1.8688*** [0.1107] 2.8782*** [0.1169]
2.0943*** [0.1150] 3.1462*** [0.1367]
Cut4 No. of observations Log likelihood Pseudo-R-squared
30082
26501
24147
21191
41427.54
36037.20
32712.35
28318.23
0.08
0.08
0.08
0.09
Robust standard errors in brackets assume clustering at country level. * significant at 10%; ** significant at 5%; *** significant at 1%. Country dummy variables included; coefficients not reported.
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
119
high-skill categories mentioned earlier (categories 3, 4, and 5) and zero otherwise. The results are a triumphant confirmation of HeckscherOhlin theory, in that the interaction term between Skill345 is negative and statistically significant. It is in fact the case that the high-skilled are more likely to support free trade in rich countries than in poor countries, just as the theory predicts. This result is also robust across specifications (although the coefficient is only at the margins of statistical significance in equation (4), with a p-value of 0.116). Equation (3) adds a variety of control variables to the regression, but the basic Heckscher-Ohlin result remains. A stated willingness to move within the country has no impact on attitudes, but international mobility is associated with free trade preferences, consistent with Rodrik (1997). Women, Roman Catholics, and older people tend toward more protectionist viewpoints, as do the unemployed. How important quantitatively is this Heckscher-Ohlin effect? Taking the specification in equation (3), and setting all right-hand side variables equal to their median values, the expected probability that a respondent will give the most protectionist response possible (protect ¼ 5) is 31.3 percent. In a country with a per capita GDP of $5000, being high-skilled reduced this probability by just 2.6 percent; but being high-skilled reduces the probability by 5.5 percent in a country with a per capita GDP of $15,000, and by 8.2 percent in a country with a per capita GDP of $25,000. It appears that income matters a lot in determining the impact of being high-skilled on preferences.8 Finally, equation (4) tests Prediction 3 by adding a measure of antiimmigrant sentiment. Prediction 3 is vindicated, in that those who are more anti-immigrant also tend to be more protectionist: trade and immigration policy are viewed as complements rather than as substitutes, just as Heckscher-Ohlin theory predicts. An alternative way of testing Heckscher-Ohlin theory is to run a series of regressions for individual countries, and see how the relationship between skills and protectionist sentiment which comes out of these regressions varies across countries. The table that makes up appendix 1 gives the result of a series of country-specific regressions, which include most of the variables in equation (3) of table 5.3.9 Figure 5.1 plots the coefficients on Skill345 for each of these countries, against that country’s GDP per capita. Again, Prediction 1 is confirmed, in that there is clearly a negative relationship between the impact of skill on protectionist attitudes, and GDP per capita. Indeed, in four of the poorest countries in the sample (Poland, Bulgaria, the Philippines, and
120
Heckscher-Ohlin Trade Theory
Figure 5.1 Impact of skill and GDP
Slovakia) being high-skilled is actually associated with being more protectionist, rather than less protectionist, although the effects are small and statistically insignificant. Ideally, of course, one would like to have information on even poorer countries, and see if skills are strongly and positively related to protectionist preferences, but this is not possible with the ISSP dataset. The robustness of these results is confirmed by Mayda and Rodrik (2005), who independently arrived at the same conclusions using slightly different methods and specifications. In particular, they
ran ordered logit rather than ordered probit regressions
used years of education rather than occupational skill-level to measure human capital
use other control variables, such as individuals’ relative incomes, and their sector of employment (which they infer from the data on occupations).
Despite these differences, their basic findings are strikingly similar to the ones presented here. Moreover, the Heckscher-Ohlin results carry over when Mayda and Rodrik employ data for a larger sample of countries (taken from the World Values Survey). It appears that people’s preferences regarding trade policy are fully consistent with the predictions of factor proportions theory.
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
121
Understanding Anti-immigrant Preferences Table 5.4 presents the results of a series of regressions explaining ‘‘antiimmigrant,’’ which is an ordered variable running from 1 (least antiimmigrant) to 5 (most anti-immigrant). The same variables are used as in the previous analysis, and as before patriotism, and especially chauvinism, are important determinants of anti-immigrant sentiment. Equation (2) tests the unconditional version of Prediction 2, and this time the results are not favorable to Heckscher-Ohlin theory: the coefficient on the interaction term between Skill345 and GDP per capita is negative, as expected, but the effect is statistically insignificant. Similarly, equation (3) tests the unconditional version of Prediction 4, and again the results are disappointing for the Borjas self-selection theory. The coefficient on an interaction term between Skill345 and the Gini coefficient is positive, as expected, but again insignificant. Equation (4) tests a conditional version of Prediction 2, and this time the results are favorable. Controlling for international differences in income distribution the interaction term between Skill345 and GDP per capita is negative and statistically significant, just as theory predicts. This result remains robust when other control variables are used (in equations 5 and 6), although the coefficient on the interaction term between Skill345 and GDP per capita becomes statistically insignificant at conventional levels in equation 6 (with a p-value of 0.113). Moreover, controlling for international income differentials, the interaction term between Skill345 and the Gini coefficient is positive in equation (4), and on the margins of statistical significance (with a p-value of 0.136). When further control variables are added to the specification in equations (5) and (6), this positive coefficient on the interaction term between Skill345 and the Gini coefficient becomes statistically significant at conventional levels, confirming a conditional version of Prediction 4. How strong are these effects? Again, taking the specification in equation (5), and setting all the explanatory variables equal to their median values, yields an expected probability of the most anti-immigrant response of 48.6 percent. Assuming that the Gini coefficient is held at its median value, 31.6, being high-skilled reduces the expected probability of the most anti-immigrant response by 3.4 percent at a per capita income of $5,000, but by 6.1 percent at per capita incomes of $15,000, and by 8.7 percent at per capita incomes of $25,000. Assuming that per capita income is held constant, at its median value
122
Heckscher-Ohlin Trade Theory
Table 5.4 Determinants of anti-immigrant preferences (ordered probit) (dependent variable: anti-immigrant) (1)
(2)
(3)
(4)
(5)
(6)
Patriotism
0.1090*** [0.0193]
0.1001*** [0.0209]
0.0988*** [0.0209]
0.0997*** [0.0208]
0.0787*** [0.0148]
0.0606*** [0.0140]
Chauvinism
0.3606*** [0.0461]
0.3415*** [0.0483]
0.3432*** [0.0483]
0.3418*** [0.0484]
0.3204*** [0.0516]
0.2833*** [0.0497]
Skill345
0.0569 [0.0705]
Skill345* GDPCAP
0.0061 [0.0047]
Skill345* Inequality
0.3366* [0.1721]
0.0056 [0.0052]
0.2662 [0.1765]
0.2784* [0.1625]
0.3045** [0.1544]
0.0072* [0.0041]
0.0068* [0.0039]
0.0060 [0.0038]
0.0070 [0.0047]
0.0072* [0.0043]
0.0083** [0.0041]
National mobility
0.0133 [0.0197]
0.0119 [0.0195]
International mobility
0.0806** [0.0326]
0.0700** [0.0338]
Never lived abroad
0.1228*** [0.0276]
0.1106*** [0.0274]
Native
0.1842*** [0.0569]
0.1860*** [0.0581]
Native parents
0.2002** [0.0779]
0.1996*** [0.0730]
Age
0.0075*** [0.0025]
0.0064*** [0.0024]
Age squared
0.0001** [0.0000]
0.0000* [0.0000]
Female
0.0327 [0.0261]
0.0096 [0.0251]
Married
0.0018 [0.0233]
0.0006 [0.0231]
Catholic
0.0230 [0.0418] 0.0370 [0.0535]
0.0281 [0.0418] 0.0284 [0.0533]
Unemployed Protectionism Cut1 Cut2
0.1228*** [0.0134] 1.0700*** 1.1802*** 1.2106*** 1.2007*** 0.7171*** 0.5098*** [0.1347] [0.1409] [0.1342] [0.1354] [0.1577] [0.1621] 0.3720*** 0.4830*** 0.5134*** 0.5032*** 0.0075 [0.1356] [0.1463] [0.1387] [0.1409] [0.1468]
0.2024 [0.1547]
Cut3
0.8796*** [0.1293]
0.7786*** [0.1420]
0.7478*** [0.1342]
0.7585*** [0.1366]
1.2907*** [0.1589]
1.5100*** [0.1643]
Cut4
1.6979*** [0.1426]
1.5894*** [0.1561]
1.5586*** [0.1485]
1.5695*** [0.1511]
2.1098*** [0.1659]
2.3363*** [0.1720]
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
123
Table 5.4 (continued) (1) No. of observations Log likelihood Pseudo-Rsquared
(2)
26484
(3)
23246
(4)
23246
(5)
23246
(6)
21220
21191
32707.20 28675.13 28676.30 28671.76 25883.56 25709.22 0.07
0.07
0.07
0.07
0.08
0.08
Robust standard errors in brackets assume clustering at country level. * significant at 10%; ** significant at 5%; *** significant at 1%. Country dummy variables included; coefficients not reported.
for this sample of countries of $19,270, being high-skilled reduces the expected probability of the most anti-immigrant response by 9.0 percent when the Gini coefficient is 25, by 6.3 percent when the Gini coefficient is 35, and by only 3.5 percent when the Gini coefficient is 45. The net impact of being high-skilled is positive for Gini coefficients of 58 and over. As before, national mobility is unrelated to attitudes to globalization, but a stated willingness to move internationally, or a history of such mobility, reduces the probability that a respondent will express antiimmigrant opinions. Both natives and the children of natives are more anti-immigrant, as are older people. In contrast with the results for trade, being a woman or a Roman Catholic does not have a statistically significant impact on preferences (and the coefficient for Roman Catholics is actually negative). Neither does being unemployed have any such effect, which is surprising. Finally, equation (6) tests Prediction 3, by including ‘‘protect’’ as an additional explanatory variable; protectionism is positively and statistically significantly correlated with anti-immigrant sentiment, just as Heckscher-Ohlin theory would predict (although, as noted above, when ‘‘protect’’ is included in the specification the interaction term between Skill345 and GDP per capita becomes statistically insignificant). Again, another approach to testing the Heckscher-Ohlin and selfselection theories is to run a series of regressions explaining attitudes towards immigration in individual countries, and compare the coefficients on Skill345 across countries. The table that makes up appendix 2 gives the results of doing this using the specification in equation (5) (without country dummies or the two interaction terms). Figure 5.2
124
Heckscher-Ohlin Trade Theory
Figure 5.2 Impact of skill and GDP
plots the resultant coefficients on Skill345 for each country, against that country’s level of GDP per capita. As can be seen, support for the HO predictions is in this case unclear. There is indeed a negative relationship between the coefficient on Skill345 and per capita GDP for the poorer countries in the sample (i.e., the Philippines and the transition economies of Central and Eastern Europe); and in two of the poorest countries, Latvia and the Philippines, the impact of skills on anti-immigrant attitudes is actually positive. However, for the richer countries in the sample the relationship is unclear. This methodology provides much stronger evidence for the Borjas theory: figure 5.3 shows a positive relationship between the Skill345 coefficient and the Gini coefficient (with a correlation coefficient of 0.401). Of course, figure 5.2 just plots the bivariate relationship between the Skill345 coefficient and GDP per capita; while the regressions in table 5.4 control for a simultaneous relationship between the Skill345 coefficient and inequality. It appears that the evidence for the predictions of Heckscher-Ohlin theory is weak when the unconditional version of that theory is tested; however, conditional on other factors the predictions of the theory hold up well. The Borjas theory does better than factor proportions theory when tested unconditionally, but does even better yet when tested conditional on other factors.
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
125
Figure 5.3 Impact of skill and inequality
The above exercises are fairly simple in their methodology. However, Mayda (2006) has recently and independently arrived at similar conclusions to these, using the same data set, as well as the World Values Survey, but going into much greater detail and employing many additional individual- and country-level variables to test the basic Heckscher-Ohlin predictions. She uses both education and skills as measures of human capital, and runs probit regressions explaining a dichotomous ‘‘immigrant opinion’’ variable. Her results are even more favorable for factor proportions theory than mine, even though she does not correct for differences in inequality across countries. The findings in this section thus appear to be robust. Explaining Attitudes Toward Trade and Immigration Simultaneously The previous sections have documented relationships between attitudes toward globalization that conform well with factor proportions theory. One objection to the results, however, is that they do not take adequate account of the fact that attitudes toward trade and immigration are correlated with each other, and (crucially) that unobserved
126
Heckscher-Ohlin Trade Theory
Table 5.5 Determinants of anti-globalization preferences (seemingly unrelated bivariate probit) (1)
(2)
Highly protectionist
Highly antiimmigrant
Patriotism
0.1967*** [0.0214]
0.0803*** [0.0225]
Chauvinism
0.3677*** [0.0285] 0.0387 [0.0717]
0.3754*** [0.0479] 0.2137 [0.1703]
0.0137*** [0.0040]
0.0093** [0.0038]
Dependent variable
Skill345 Skill345*GDPCAP Skill345*Inequality National mobility International mobility Never lived abroad Native Native parents Age Age squared Female
0.0057 [0.0040] 0.0301 [0.0189]
0.0063 [0.0176]
0.0029 [0.0324] 0.0330 [0.0310]
0.0233 [0.0292] 0.0537 [0.0363]
0.0873 [0.0827]
0.2182** [0.0873]
0.0466 [0.0690]
0.2515*** [0.0785]
0.0164*** [0.0049]
0.0204*** [0.0031]
0.0001*** [0.0000] 0.0985*** [0.0262]
0.0002*** [0.0000] 0.0301 [0.0224]
Married
0.0086 [0.0194]
0.0239 [0.0231]
Catholic
0.0588*** [0.0226]
0.0082 [0.0293]
Unemployed
0.0917** [0.0362]
0.0986* [0.0580]
Constant
2.8535*** [0.1675]
2.7675*** [0.1754]
No. of observations
24180
Rho [standard error of rho]
0.221349 [0.013959]
Wald test of rho ¼ 0
Chi-squaredð1Þ ¼ 235.13, p-value ¼ 0.0000
Robust standard errors in brackets assume clustering at country level. * significant at 10%; ** significant at 5%; *** significant at 1%. Country dummy variables included; coefficients not reported.
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
127
determinants of globalization could have similar effects on both variables. Table 5.5 therefore presents the results of seemingly unrelated bivariate probit regressions explaining attitudes toward both trade and immigration. It estimates two regressions with the same explanatory variables as before, but allows the disturbance terms in both regressions to be correlated with each other.10 The dependent variables in both cases are binary variables, indicating whether the respondent gave the most anti-globalization response possible: ‘‘Highly protectionist’’ is 1 if ‘‘protect’’ ¼ 5, while ‘‘Highly anti-immigrant’’ is 1 if ‘‘anti-immigrant’’ ¼ 5; otherwise both variables are zero. The ‘‘rho’’ coefficient reported at the bottom is the correlation between the disturbances in the two equations, or ‘‘(roughly) the correlation between the outcomes after the influence of the included factors is accounted for’’ (Greene 2000:854). The results confirm Prediction 3 in that ‘‘rho’’ is strongly positive. Predictions 1 and 2 are also confirmed, in that the interaction terms between Skill345 and GDP per capita are negative. There is less strong support for Prediction 4: while the interaction term between Skill345 and the Gini coefficient in equation (2) is positive, it is statistically insignificant at conventional levels (with a p-value of 0.151). The other big difference between the results here and those obtained earlier is that unemployment now has a positive effect on anti-immigrant sentiment, as well as on protectionism, while the international mobility variables are no longer statistically significant. Conclusion Presumably, the debate about to what extent factor proportions theory explains trade flows will continue in the decades ahead. However, it appears that people’s attitudes toward globalization are strikingly similar to those that would be predicted if Heckscher-Ohlin trade theory accurately described the world. The high-skilled are proglobalization in rich countries, confirming the results of Scheve and Slaughter. Even more tellingly, in some of the very poorest countries in the ISSP sample, being high-skilled has a negative (if statistically insignificant) impact on pro-globalization sentiment. More generally, an interaction term between skills and GDP per capita has a negative impact in regressions explaining antiglobalization sentiment. Furthermore, individuals view protectionism and anti-immigrant policies as complements rather than as substitutes, which is what simple HeckscherOhlin theory predicts.
128
Heckscher-Ohlin Trade Theory
Appendix 1. Country-specific regressions: Protectionism
Patriotism Chauvinism Skill345 National mobility
(1)
(2)
(3)
(4)
(5)
(6)
Australia
West Germany
East Germany
Great Britain
USA
Austria
0.2555*** [0.0446] 0.3432*** [0.0368]
0.3298*** [0.0782] 0.3975*** [0.0682]
0.2087*** 0.2394*** 0.3260** [0.0531] [0.0792] [0.1269] 0.0235 [0.0555]
International 0.1063 mobility [0.0707] Never lived abroad Native
0.1199** [0.0508] 0.4638*** [0.0466]
0.1358** [0.0618] 0.0804 [0.1517]
0.2382*** 0.3146*** [0.0726] [0.1079] 0.2263*** 0.0530 [0.0870] [0.1604]
0.1764*** [0.0590] 0.4947*** [0.0558]
0.2807*** [0.0572] 0.4284*** [0.0464]
0.0536 [0.0559] 0.4504*** [0.0520]
0.3403*** 0.3248*** 0.2427*** [0.0822] [0.0646] [0.0940] 0.0942 [0.0839]
0.0487 [0.0762]
0.1706** [0.0841]
0.1316 [0.0946]
0.0031 [0.0925]
0.1257 [0.1088]
0.1356 [0.0956] 0.1661 [0.2457]
0.3172 [0.2088] 1.5934* [0.8838]
0.0415 [0.0930] 0.0019 [0.2363]
0.0682 [0.0835] 0.1027 [0.2778]
0.1329 [0.1085] 0.5752** [0.2689]
Native parents
0.1709 [0.1401]
0.2020 [0.2223]
1.1386 [0.7053]
0.1710 [0.2337]
0.0563 [0.2653]
0.6279** [0.2465]
Age
0.0159 [0.0119]
0.0147 [0.0131]
0.0106 [0.0205]
0.0254** [0.0128]
0.0286** [0.0113]
0.0099 [0.0116]
0.0002 [0.0001]
0.0002 [0.0001]
0.0001 [0.0002]
0.0003** [0.0001]
0.0001 [0.0001]
Age squared Female Married
0.3258*** 0.3708*** [0.0522] [0.0668] 0.0235 0.1244 [0.0634] [0.0789]
0.0002* [0.0001]
0.6695*** 0.2217*** [0.1010] [0.0755] 0.0247 0.0310 [0.1160] [0.0777]
0.1663*** [0.0634] 0.0839 [0.0639]
0.3016*** [0.0720] 0.1429* [0.0796]
0.0695 [0.0608]
0.0535 [0.0670]
0.0606 [0.2462]
0.1018 [0.1176]
0.0142 [0.0720]
0.1011 [0.0910]
Unemployed
0.0572 [0.1382]
0.0887 [0.1957]
0.2491 [0.1553]
0.2157 [0.1599]
0.1846 [0.1864]
0.2144 [0.1736]
Cut1
0.7244** [0.3403]
0.3004 [0.3233]
0.6345 [0.6688]
0.5813 [0.3619]
1.1999*** [0.3562]
0.0154 [0.3222]
Cut2
0.4602 [0.3263]
1.4560*** [0.3247]
1.7858*** [0.6661]
1.8647*** [0.3469]
2.1036*** [0.3574]
0.9426*** [0.3209]
Cut3
0.9978*** [0.3274]
2.1714*** [0.3276]
2.5158*** [0.6701]
2.6709*** [0.3517]
2.8997*** [0.3616]
1.4623*** [0.3235]
Cut4
2.2817*** [0.3301]
3.1295*** [0.3332]
3.5303*** [0.6753]
3.9466*** [0.3615]
4.2290*** [0.3708]
2.4259*** [0.3289]
Catholic
No. of observations
1877
1067
512
923
1225
985
Log likelihood Pseudo-Rsquared
2212.59
1480.28
664.43
1139.25
1530.54
1251.86
0.07
0.10
0.14
0.10
0.09
0.09
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
Patriotism Chauvinism
(7)
(8)
(9)
(10)
(11)
(12)
Hungary
Italy
Ireland
Netherlands
Norway
Sweden
0.0551 [0.0542] 0.2131*** [0.0476]
Skill345
0.0455 [0.0832]
National mobility
0.0302 [0.0842]
International 0.1429 mobility [0.1267] Never lived abroad Native
129
0.0184 [0.1599] 0.8959** [0.3914]
0.1766*** [0.0493] 0.2984*** [0.0541]
0.2146*** [0.0575] 0.3735*** [0.0579]
0.0781* [0.0400] 0.4806*** [0.0395]
0.2171*** [0.0522] 0.2797*** [0.0437]
0.3390*** [0.0520] 0.3575*** [0.0469]
0.2143** [0.0857]
0.1716*** 0.2049*** [0.0540] [0.0614]
0.0660 [0.0727]
0.0353 [0.0832]
0.0325 [0.0551]
0.0931 [0.0639]
0.0231 [0.0893]
0.1055 [0.1072]
0.0973 [0.0675]
0.2835*** 0.2649*** [0.0833] [0.0847]
0.2647** [0.1078] 0.1237 [0.5965]
0.1836** [0.0811] 0.4066 [0.2558]
0.1485** [0.0736] 0.1089 [0.2538]
0.1604** [0.0742] 0.2799 [0.2240]
0.1104 [0.0710]
0.1952** [0.0852] 0.4316 [0.3256]
Native parents
0.3273 [0.2724]
0.0242 [0.2703]
0.3261* [0.1917]
0.1300 [0.2327]
0.2287 [0.1932]
0.6031* [0.3105]
Age
0.0062 [0.0122]
0.0143 [0.0148]
0.0111 [0.0138]
0.0103 [0.0093]
0.0081 [0.0105]
0.0118 [0.0130]
Age squared
0.0000 [0.0001]
0.0002 [0.0002]
0.0001 [0.0001]
0.0001 [0.0001]
0.0001 [0.0001]
0.0001 [0.0001]
Female
0.0656 [0.0735] 0.0063 [0.0757]
Married
0.2150*** 0.3492*** [0.0659] [0.0730] 0.1162 0.0776 [0.0815] [0.0829]
0.3406*** [0.0506] 0.0852 [0.0605]
0.2370*** 0.5068*** [0.0577] [0.0638] 0.0588 0.0739 [0.0695] [0.0731]
0.0219 [0.0776]
0.1304 [0.1710]
0.0698 [0.1343]
0.0584 [0.0624]
0.9099** [0.4175]
0.1306 [0.1418]
0.0745 [0.1792]
0.2189* [0.1273]
0.0491 [0.1251]
0.1349 [0.1483]
0.2740** [0.1245]
Cut1
1.4153*** [0.4183]
0.5839 [0.6489]
0.3938 [0.4560]
0.3055 [0.2599]
0.2325 [0.3207]
0.8668** [0.3583]
Cut2
0.7563* [0.4138]
1.4071** [0.6476]
0.9237** [0.4419]
1.7456*** [0.2603]
0.9902*** [0.3185]
1.8561*** [0.3574]
Cut3
0.0136 [0.4110] 0.7238* [0.4111]
1.9248*** [0.6479] 2.8932*** [0.6488]
1.3130*** [0.4431] 2.5078*** [0.4463]
2.6989*** [0.2623] 3.9858*** [0.2709]
1.9691*** [0.3204] 3.1494*** [0.3264]
2.9802*** [0.3637] 4.0949*** [0.3734]
Catholic Unemployed
Cut4
0.3315 [0.2897]
No. of observations
983
1084
942
1853
1391
1186
Log likelihood
1251.18
1553.23
1221.08
2363.53
1844.13
1540.38
Pseudo-Rsquared
0.02
0.04
0.06
0.07
0.07
0.11
130
Patriotism Chauvinism
Heckscher-Ohlin Trade Theory
(13)
(14)
(15)
(16)
(17)
(18)
Czech Rep.
Slovenia
Poland
Bulgaria
Russia
New Zealand
0.1547*** [0.0534] 0.3188*** [0.0496]
0.2286*** [0.0544] 0.3749*** [0.0538]
0.2071*** [0.0600] 0.2448*** [0.0597]
0.1000* [0.0517] 0.3817*** [0.0551]
0.2281*** [0.0402] 0.2932*** [0.0388]
0.0222 [0.0794]
0.1294 [0.0894]
0.1738** [0.0728]
0.1223* [0.0631] 0.4073*** [0.0521]
Skill345
0.1966*** 0.4022*** [0.0729] [0.0838]
National mobility
0.0649 [0.0740]
0.0985 [0.0780]
0.0263 [0.0715]
0.0296 [0.0856]
0.1257* [0.0709]
0.0346 [0.0808]
International 0.1304 mobility [0.1167]
0.0340 [0.1209]
0.0222 [0.0885]
0.2124** [0.1004]
0.0401 [0.0995]
0.0059 [0.0920]
Never lived abroad Native
0.0703 [0.1133] 0.6694* [0.3824]
0.1370 [0.0851] 0.3807** [0.1882]
0.1406 [0.1188] 0.6276* [0.3690]
0.1380 [0.1066] 0.0129 [0.4177]
0.1945 [0.1458] 0.5217 [0.3919]
0.1621* [0.0839] 0.2263 [0.2239]
0.1537 [0.2525]
0.1932 [0.1671]
0.6299** [0.2523]
0.7981*** 0.1941 [0.2394] [0.3528]
0.1369 [0.2065]
0.0031 [0.0134]
0.0186 [0.0138]
0.0113 [0.0126]
0.0276** [0.0133]
0.0027 [0.0141]
Age squared
0.0001 [0.0001]
0.0002 [0.0001]
0.0002 [0.0001]
0.0002* [0.0001]
Female
0.2132*** 0.1194* [0.0692] [0.0710] 0.0694 0.0053 [0.0803] [0.0812]
0.0244 [0.0697] 0.0713 [0.0826]
0.0169 [0.0739] 0.0705 [0.0878]
0.0790 [0.0567] 0.1401** [0.0613]
0.2428*** [0.0725] 0.1726** [0.0843]
Native parents Age
Married
0.0284*** [0.0101] 0.0002 [0.0001]
0.2421*** [0.0827]
0.0000 [0.0001]
0.0933 [0.0734]
0.0747 [0.0855]
0.0614 [0.1025]
0.6094*** [0.1692]
0.0478 [1.0353]
0.0675 [0.1089]
Unemployed
0.0565 [0.2562]
0.0549 [0.1308]
0.0915 [0.1404]
0.1030 [0.1172]
0.0663 [0.1117]
0.0226 [0.1867]
Cut1
0.1467 [0.4185]
0.4850 [0.3470]
0.1140 [0.4521]
1.6706*** [0.5222]
1.2945*** [0.3445]
0.0797 [0.4284]
Cut2
0.6079 [0.4191]
0.7362** [0.3388]
0.7941* [0.4528]
1.9691*** [0.5197]
2.0614*** [0.3423]
1.1461*** [0.4187]
Cut3
1.2235*** [0.4211] 2.0054*** [0.4233]
1.4911*** [0.3422] 2.3164*** [0.3460]
1.5049*** [0.4545] 2.4589*** [0.4574]
2.6696*** [0.5225] 3.4136*** [0.5250]
2.6419*** [0.3448] 3.3397*** [0.3481]
1.7926*** [0.4196] 2.8628*** [0.4217]
Catholic
Cut4 No. of observations
994
993
1043
1050
1566
893
Log likelihood
1458.69
1370.44
1446.36
1184.65
2195.17
1251.33
Pseudo-Rsquared
0.06
0.07
0.03
0.06
0.06
0.06
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
Patriotism Chauvinism Skill345 National mobility
131
(19)
(20)
(21)
(22)
(23)
(24)
Canada
Philippines
Japan
Spain
Latvia
Slovakia
0.0908** [0.0447] 0.3879*** [0.0457] 0.0678 [0.0695] 0.1580** [0.0767]
0.1929*** [0.0572] 0.1903*** [0.0626]
0.1441*** [0.0500] 0.3015*** [0.0400]
0.1269** [0.0550] 0.2956*** [0.0619] 0.0578 [0.1019]
0.0055 [0.1654]
0.1048** [0.0477] 0.2569*** [0.0497]
0.2921*** [0.0449] 0.1678*** [0.0404]
0.0494 [0.1027]
0.0559 [0.0743]
0.1235* [0.0719]
0.0061 [0.0694]
0.1569** [0.0726]
0.0121 [0.0894]
0.0271 [0.0643]
0.1248 [0.0765]
0.1388 [0.1269]
0.0937 [0.0965]
0.0213 [0.1166]
0.1289* [0.0765]
0.0458 [0.0787] 0.2452 [0.1626]
0.0118 [0.1171] 0.5194 [0.3208]
0.1447 [0.1487]
0.1946* 0.0255 [0.1074] [0.1042] 0.3537*** 0.0017 [0.0976] [0.1354]
0.1228 [0.1011] 0.4517 [0.3207]
Native parents
0.1970 [0.1428]
0.4163* [0.2182]
0.3030 [0.2244]
0.3334 [0.2699]
0.2176* [0.1247]
0.2845 [0.2072]
Age
0.0086 [0.0123]
0.0005 [0.0116]
0.0541*** [0.0113]
0.0116 [0.0108]
0.0244 [0.0154]
0.0152 [0.0118]
Age squared
0.0001 [0.0001]
0.0000 [0.0001]
0.0005*** 0.0002 [0.0001] [0.0001]
0.0002 [0.0002]
0.0001 [0.0001]
Female
0.2290*** 0.0045 [0.0633] [0.0637] 0.0760 0.0617 [0.0714] [0.0755]
0.4068*** [0.0628] 0.1528* [0.0860]
0.1372** [0.0658] 0.0892 [0.0809]
0.0664 [0.0786] 0.0918 [0.0808]
0.1365** [0.0594] 0.0006 [0.0683]
0.4218*** [0.1240]
0.1623 [0.1013]
0.0851 [0.0611]
International 0.0961 mobility [0.0804] Never lived abroad Native
Married Catholic
0.0195 [0.0652]
0.0529 [0.0920]
0.4129 [0.5195]
Unemployed
0.4242** [0.1828]
0.2199 [0.1582]
0.1006 [0.3267]
0.1069 [0.1091]
0.0133 [0.1253]
0.1245 [0.1239]
1.9101*** [0.4886]
0.0324 [0.3559]
0.1656 [0.4003]
0.0510 [0.3569]
0.2988 [0.3721]
0.4834 [0.4759]
0.5485 [0.3576]
1.3108*** [0.4099]
0.6901* [0.3542]
1.0663*** [0.3739]
1.5325*** [0.3596] 2.1536*** [0.3624]
2.0450*** [0.4106] 3.4891*** [0.4165]
1.2806*** [0.3593] 1.8672*** [0.3615]
1.6687*** [0.3762] 2.4750*** [0.3789]
Cut1
0.1839 [0.3475]
Cut2
0.8776** [0.3463]
Cut3
1.6362*** [0.3490] 2.6894*** [0.3530]
Cut4
0.0950 [0.4757] 1.7006*** [0.4790]
No. of observations
1149
1197
1228
1184
974
1346
Log likelihood
1631.23
1444.71
1806.41
1420.20
1230.11
1967.98
Pseudo-Rsquared
0.04
0.02
0.05
0.04
0.04
0.05
Note: Robust standard errors in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%.
132
Heckscher-Ohlin Trade Theory
Appendix 2. Country-specific regressions: Anti-immigration
Patriotism Chauvinism Skill345 National mobility
(1)
(2)
(3)
(4)
(5)
(6)
Australia
West Germany
East Germany
Great Britain
USA
Austria
0.1691*** [0.0432] 0.4498*** [0.0358]
0.1265** [0.0548] 0.5690*** [0.0588]
0.1589* [0.0833] 0.4891*** [0.0764]
0.1802*** 0.2595*** 0.1637 [0.0521] [0.0931] [0.1359] 0.0224 [0.0555]
International 0.0030 mobility [0.0692]
0.0990 [0.0674] 0.5841*** [0.0616]
0.1471** [0.0579] 0.2095*** [0.0499]
0.0634 [0.0544] 0.4231*** [0.0514]
0.1326 [0.0857]
0.0748 [0.0676]
0.1088 [0.0992]
0.0067 [0.0849]
0.2066** [0.0855]
0.0363 [0.0871]
0.0700 [0.1231]
0.0544 [0.0887]
0.2028** [0.0957]
0.1327 [0.1935]
0.1020 [0.1084]
0.3382 [0.2487] 0.3647 [0.7416]
0.0314 [0.0976] 0.0397 [0.2545]
0.3167*** 0.2189 [0.1008] [0.1362]
Never lived abroad Native
0.2430*** 0.0019 [0.0609] [0.1107] 0.0029 0.1572 [0.1328] [0.2965]
Native parents
0.0076 [0.1227]
0.4100 [0.2581]
0.0633 [0.6680]
0.1442 [0.2566]
0.6886*** [0.2241]
0.2090 [0.2570]
0.0158 [0.0112]
0.0198 [0.0145]
0.0120 [0.0236]
0.0229* [0.0132]
0.0177 [0.0122]
0.0171 [0.0134]
Age squared
0.0001 [0.0001]
0.0002 [0.0001]
0.0001 [0.0003]
0.0002* [0.0001]
0.0002 [0.0001]
0.0002 [0.0001]
Female
0.2092*** [0.0519] 0.0659 [0.0632]
0.0680 [0.0783] 0.1019 [0.0935]
0.0099 [0.1118] 0.1030 [0.1361]
0.1179 [0.0802] 0.1176 [0.0835]
0.1091 [0.0686] 0.0198 [0.0691]
0.0283 [0.0742] 0.0454 [0.0879]
0.2299*** [0.0639]
0.0076 [0.0774]
0.1096 [0.2781]
0.1125 [0.1231]
0.1233 [0.0754]
0.1587 [0.0990]
0.0093 [0.2047]
0.4226 [0.2809]
0.1199 [0.1963]
0.0102 [0.1775]
0.2691 [0.2093]
0.3207 [0.1979]
Cut1
0.6585** [0.3214]
0.2309 [0.4137]
1.1444* [0.6673]
0.2445 [0.3795]
0.0459 [0.3831]
0.8801** [0.3892]
Cut2
0.2559 [0.3163]
0.6449* [0.3734]
0.6495 [0.6349]
0.8527** [0.3865]
0.5854 [0.3856]
0.1508 [0.3699]
Cut3
1.3103*** [0.3186]
2.0472*** [0.3683]
0.7616 [0.6215]
2.3958*** [0.3963]
1.6754*** [0.3945]
1.6863*** [0.3726]
Cut4
2.2328*** [0.3212]
2.8951*** [0.3737]
1.5257** [0.6265]
3.1593*** [0.4011]
2.5030*** [0.3998]
2.4364*** [0.3760]
Age
Married Catholic Unemployed
0.1171 [0.0948] 0.2970 [0.2391]
0.0214 [0.1147] 0.1517 [0.2823]
No. of observations
1831
963
478
870
1074
927
Log likelihood Pseudo-Rsquared
2322.91
970.49
462.17
959.05
1381.60
1061.30
0.08
0.13
0.09
0.10
0.05
0.07
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
Patriotism Chauvinism
133
(7)
(8)
(9)
(10)
(11)
(12)
Hungary
Italy
Ireland
Netherlands
Norway
Sweden
0.0786 [0.0588] 0.0909* [0.0513]
0.0861* 0.0342 [0.0490] [0.0604] 0.3801*** 0.2621*** [0.0545] [0.0595]
0.0822 [0.0536] 0.7179*** [0.0495]
0.1860** [0.0881]
National mobility
0.0275 [0.0890]
0.0722 [0.0795]
0.2395*** 0.0418 [0.0868] [0.0598]
0.0327 [0.0672]
0.0017 [0.0755]
International mobility
0.0242 [0.1422]
0.0076 [0.0899]
0.0514 [0.1164]
0.1015 [0.0688]
0.2068** [0.0926]
0.1199 [0.0860]
Never lived abroad Native
0.1301 [0.1721] 0.4473 [0.4180]
0.0751 [0.1189] 0.1864 [0.4917]
0.1356 [0.0833] 0.2027 [0.3183]
0.0093 [0.0801] 0.0823 [0.2817]
0.0838 [0.0829] 0.3568* [0.2144]
0.0153 [0.0952] 0.6224** [0.2758]
0.0329 [0.3814]
0.6629 [0.4079]
0.0064 [0.2273]
0.5540** [0.2314]
0.2391 [0.1512]
0.6648*** [0.2473]
Age
0.0015 [0.0124]
0.0101 [0.0158]
0.0242 [0.0149]
0.0052 [0.0103]
0.0072 [0.0113]
0.0088 [0.0140]
Age squared
0.0000 [0.0001]
0.0001 [0.0002]
0.0003* [0.0001]
0.0000 [0.0001]
0.0001 [0.0001]
0.0002 [0.0001]
Female
0.1316* [0.0787] 0.0267 [0.0802]
0.0808 [0.0713] 0.0776 [0.0881]
0.0160 [0.0750] 0.0646 [0.0872]
0.0123 0.0850 [0.0534] [0.0616] 0.1711*** 0.1830** [0.0652] [0.0752]
0.0205 [0.0676] 0.0413 [0.0762]
0.0222 [0.0797]
0.0705 [0.1558]
0.4111** [0.1744]
0.1286* [0.0661]
0.1722 [0.8406]
0.7988* [0.4165]
0.1279 [0.1434]
0.0711 [0.2171]
0.0742 [0.1375]
0.1916 [0.1485]
0.1571 [0.1823]
0.0086 [0.1165]
Married Catholic Unemployed
0.1249** [0.0580]
0.1340** [0.0533] 0.6451*** [0.0465]
Skill345
Native parents
0.0084 [0.0832]
0.0331 [0.0423] 0.6969*** [0.0438]
0.2383*** [0.0659]
Cut1
1.2354*** 1.8930*** 1.5440*** [0.4452] [0.4957] [0.5401]
0.1337 [0.3170]
0.1577 [0.3268]
0.2074 [0.3848]
Cut2
0.9748** [0.4370]
1.2968*** 0.4081 [0.4949] [0.5339]
1.0221*** [0.2985]
1.1489*** [0.3170]
0.7844** [0.3851]
Cut3
0.2251 [0.4422] 1.0080** [0.4466]
0.1063 [0.4947] 0.8108 [0.4958]
2.6047*** [0.3013] 3.6460*** [0.3061]
2.5250*** [0.3212] 3.5019*** [0.3247]
1.9677*** [0.3864] 2.9487*** [0.3916]
Cut4
1.3235** [0.5381] 2.0470*** [0.5440]
No. of observations
937
1033
885
1744
1311
1105
Log likelihood
939.29
1169.78
1004.79
1962.22
1515.48
1275.69
Pseudo-Rsquared
0.01
0.05
0.03
0.12
0.13
0.12
134
Patriotism Chauvinism Skill345
Heckscher-Ohlin Trade Theory
(13)
(14)
(15)
(16)
(17)
(18)
Czech Rep.
Slovenia
Poland
Bulgaria
Russia
New Zealand
0.1340** [0.0532] 0.1895*** [0.0536]
0.1365** [0.0590] 0.3389*** [0.0584]
0.2979*** 0.2013** [0.0805] [0.0859]
0.1544** [0.0671] 0.1338** [0.0682]
0.0786 [0.0554] 0.0030 [0.0666]
0.2489** [0.0975]
0.0252 [0.1012]
0.0562 [0.0464] 0.1177** [0.0484] 0.0819 [0.0853]
0.1622*** [0.0627] 0.3535*** [0.0531] 0.2602*** [0.0883]
0.1201 [0.0809]
0.1017 [0.0846]
0.0107 [0.0908]
0.3727*** [0.1060]
0.0729 [0.0881]
0.0163 [0.0855]
International 0.1818 mobility [0.1399]
0.0830 [0.1225]
0.1208 [0.1144]
0.2454** [0.1187]
0.0381 [0.1067]
0.2254** [0.0928]
0.1228 [0.1266] 0.0004 [0.3340]
0.0812 [0.0976] 0.2128 [0.2691]
0.2199 [0.1387] 0.0964 [0.3124]
0.0582 [0.1751] 0.2104 [0.2758]
0.1516* [0.0864] 0.0660 [0.1790]
Native parents
0.0567 [0.2680]
0.4725* [0.2545]
0.1278 [0.2438]
0.6822** [0.2762]
0.0877 [0.2276]
0.2620* [0.1578]
Age
0.0120 [0.0150]
0.0088 [0.0161]
0.0074 [0.0148]
0.0154 [0.0179]
0.0050 [0.0126]
0.0012 [0.0144]
Age squared
0.0001 [0.0002]
0.0001 [0.0002]
0.0000 [0.0002]
0.0001 [0.0002]
0.0001 [0.0001]
0.0001 [0.0001]
Female
0.0534 [0.0769] 0.1638* [0.0917]
0.1809** [0.0736] 0.0041 [0.0973]
0.2130** [0.0840] 0.0352 [0.0991]
0.2781*** [0.0917] 0.0434 [0.1128]
0.1357** [0.0677] 0.0112 [0.0728]
0.0276 [0.0744] 0.0024 [0.0922]
Catholic
0.0630 [0.0809]
0.1290 [0.0916]
0.0175 [0.1219]
1.7180*** [0.6228]
0.2830 [0.2286]
0.3312*** [0.1084]
Unemployed
0.1091 [0.3196]
0.3149** [0.1358]
0.0932 [0.1559]
0.4441*** [0.1616]
0.2550** [0.1215]
0.3429* [0.1931]
Cut1
1.6894*** 0.9011** [0.4608] [0.4053]
0.2647 [0.4545]
0.0772 [0.5203]
Cut2
0.9684** [0.4264]
0.3146 [0.4513]
0.3136 [0.5037]
1.4038*** [0.4560] 2.1029*** [0.4589]
1.1781** [0.5015] 1.9981*** [0.5033]
National mobility
Never lived abroad Native
Married
Cut3 Cut4
0.4433 [0.4312] 1.2302*** [0.4342]
0.3209 [0.3966] 1.5952*** [0.4029] 2.5002*** [0.4067]
0.0297 [0.1355] 0.4477 [0.3495]
1.1187*** 0.5046 [0.4171] [0.4168] 0.4916 [0.4123] 0.7046* [0.4134] 1.6143*** [0.4153]
0.3382 [0.4084] 1.3349*** [0.4081] 2.2716*** [0.4098]
No. of observations
886
932
718
672
1031
848
Log likelihood
992.49
1028.82
931.24
753.32
1360.55
1095.52
Pseudo-Rsquared
0.03
0.06
0.04
0.04
0.01
0.08
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
Patriotism Chauvinism
135
(19)
(20)
(21)
(22)
(23)
(24)
Canada
Philippines
Japan
Spain
Latvia
Slovakia
0.0562 0.0889* [0.0489] [0.0513] 0.4963*** 0.1309** [0.0502] [0.0572]
Skill345
0.1985*** [0.0751]
0.0834 [0.1556]
National mobility
0.0289 [0.0820]
0.0543 [0.0701]
International mobility
0.0106 [0.0840]
Never lived abroad Native
0.2446*** [0.0792] 0.3823** [0.1645]
0.2460*** [0.0582] 0.1261*** [0.0450]
0.1525** [0.0763]
0.2241*** 0.0058 [0.0790] [0.1367] 0.2315* [0.1187] 0.1799 [0.2733]
0.0649 [0.0539] 0.1415** [0.0588]
0.0313 [0.0570] 0.2737*** [0.0581]
0.0284 [0.0506] 0.1093** [0.0426]
0.0449 [0.1099]
0.0744 [0.1091]
0.0930 [0.0802]
0.0344 [0.0817]
0.0483 [0.1011]
0.1247* [0.0719]
0.0264 [0.0939]
0.1233 [0.1339]
0.0101 [0.0893]
0.5870*** 0.1017 [0.1416] [0.1084] 0.0818 [0.1426]
0.1460 [0.1117] 0.5318*** [0.1630]
0.1228 [0.1167] 0.7441*** [0.2658]
0.2197 [0.1463]
0.2215 [0.2722]
0.2651* [0.1445]
0.3677** [0.1804]
0.5376*** 0.0520 [0.1372] [0.1569]
0.0110 [0.0126]
0.0126 [0.0129]
0.0023 [0.0132]
0.0182 [0.0125]
0.0221 [0.0183]
0.0222* [0.0133]
Age squared
0.0001 [0.0001]
0.0001 [0.0001]
0.0001 [0.0001]
0.0001 [0.0001]
0.0001 [0.0002]
0.0002 [0.0001]
Female
0.0876 [0.0687] 0.0100 [0.0745]
0.0773 [0.0642] 0.0137 [0.0802]
0.3401*** 0.0379 [0.0678] [0.0692] 0.0660 0.0845 [0.0971] [0.0859]
0.1101 [0.0913] 0.1132 [0.0968]
0.0038 [0.0667] 0.0474 [0.0787]
0.2589*** [0.0724]
0.1281 [0.0909]
0.9550 [0.5857]
0.2370* [0.1351]
0.0848 [0.1110]
0.1233* [0.0689]
0.1409 [0.2236]
0.1958 [0.1881]
0.1018 [0.2741]
0.0173 [0.1147]
0.1497 [0.1408]
0.1495 [0.1564]
Cut1
0.1012 [0.3762]
1.4339*** [0.4578]
0.6026* [0.3466]
0.7925** [0.3760]
0.6815 [0.4748]
0.5833 [0.3766]
Cut2
0.6242* [0.3711]
0.8690* [0.4544]
1.4900*** [0.3525]
0.0852 [0.3657]
1.7902*** 0.0320 [0.4396] [0.3708]
Cut3
1.8080*** [0.3743] 2.5966*** [0.3794]
0.0535 [0.4555] 0.7908* [0.4567]
2.8113*** [0.3596] 3.6732*** [0.3650]
1.7354*** [0.3744] 2.8048*** [0.3791]
2.4255*** [0.4437]
Native parents Age
Married Catholic Unemployed
Cut4
1.4411*** [0.3703] 2.2202*** [0.3716]
No. of observations
1009
1144
1024
1045
813
1102
Log likelihood
1380.09
1586.02
1335.87
1211.95
758.86
1318.47
Pseudo-Rsquared
0.06
0.01
0.06
0.01
0.12
0.02
Note: Robust standard errors in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%.
136
Heckscher-Ohlin Trade Theory
Acknowledgments This essay was in part written while the author was an IRCHSS Government of Ireland Senior Fellow. I thank the Irish Research Council for the Humanities and Social Sciences for its generous support. I am extremely grateful to Richard Sinnott for allowing me to draw on our joint work, and I also wish to thank Kevin Denny, Chris Minns, and participants at the Eli Heckscher Celebratory Symposium for helpful suggestions. The usual disclaimer applies. Notes 1. Since beginning this project, Sinnott and I became aware of the independent work that was being done on the same issues by Anna Maria Mayda and Dani Rodrik (Mayda and Rodrik 2005; Mayda 2006). While their findings confirm our own, they use different methods and measures of skill, and the essay will allude to these differences when our own empirical findings are discussed. 2. Furthermore, it is no longer the case that trade and factor flows are necessarily substitutes: they could instead be complements. For example, Markusen (1983) shows that technological differences between countries can lead to trade and factor mobility being complements; while in the context of a three-factor model such as the specific-factors model, trade, and factor mobility can be either substitutes or complements (O’Rourke and Williamson 1999). 3. In principle, self-selection should depend not only on income distribution within host countries, but also on the relationship between host-country and source-country income distribution. A complete test of the Borjas theory would thus involve calculating source country distributions for each host country. In this essay I make the simplifying assumption that source-country distributions are sufficiently similar for all host countries that self-selection varies across host countries based on differences in host-country distributions alone. 4. The next section draws on O’Rourke and Sinnott (2001). 5. Details available on request. 6. The data are available online at http://devdata.worldbank.org/dataonline/. 7. Note that in the context of an ordered probit model, a significant positive coefficient indicates that increasing the relevant independent variable increases the probability that ‘‘protect’’ takes on the value 5, and reduces the probability that ‘‘protect’’ takes on the value 1. The impact on the probabilities that ‘‘protect’’ takes on the values 2–4 is, however, a priori unclear. Nonetheless, in what follows I will speak loosely of variables being either positively or negatively related to antiglobalization sentiment. See Greene (2000:875–879) for further details. 8. These results, and similar ones quoted in the next section, were calculated using the CLARIFY programme described in Tomz, Wittenberg, and King (1999) and King, Tomz, and Wittenberg (2000).
Heckscher-Ohlin Theory and Individual Attitudes Toward Globalization
137
9. Country dummy variables are obviously omitted from these regressions, as is the interaction term between Skill345 and GDP. 10. See Greene (2000:849–856).
References Barro, R. J., and J.-W. Lee. 2000. ‘‘International Data on Educational Attainment: Updates and Implications.’’ Harvard Center for International Development Working Paper No. 42. Available at http://www.cid.harvard.edu/ciddata/ciddata.html. Borjas, G. J. 1987. ‘‘Self-Selection and the Earnings of Immigrants.’’ American Economic Review 77:531–553. Bowen, H. P., E. E. Leamer, and L. Sveikauskas. 1987. ‘‘Multicountry, Multifactor Tests of the Factor Abundance Theory.’’ American Economic Review 77:791–809. Davis, D. R., and D. E. Weinstein. 2001. ‘‘An Account of Global Factor Trade.’’ American Economic Review 91:1423–1453. Greene, W. H. 2000. Econometric Analysis. London: Prentice-Hall. ILO. 1990. International Standard Classification of Occupations: ISCO-88. Geneva: International Labour Organization. Katz, E., and O. Stark. 1984. ‘‘Migration and Asymmetric Information: Comment.’’ American Economic Review 74:533–534. ———. 1987. ‘‘International Migration Under Asymmetric Information.’’ Economic Journal 97:718–726. King, G., M. Tomz, and J. Wittenberg. 2000. ‘‘Making the Most of Statistical Analyses: Improving Interpretation and Presentation.’’ American Journal of Political Science 44:341– 355. Leontief, W. 1953. ‘‘Domestic Production and Foreign Trade: The American Capital Position Re-Examined.’’ Proceedings of the American Philosophical Society 97:332–349. Markusen, J. R. 1983. ‘‘Factor Movements and Commodity Trade as Complements.’’ Journal of International Economics 13:341–356. Mayda, A. M. 2006. ‘‘Who Is Against Immigration? A Cross-Country Investigation of Individual Attitudes Towards Immigrants.’’ Review of Economics and Statistics, forthcoming. Mayda, A. M., and D. Rodrik. 2005. ‘‘Why Are Some People (and Countries) More Protectionist than Others?’’ European Economic Review 49: 1393–1691. Mundell, R. A. 1957. ‘‘International Trade and Factor Mobility.’’ American Economic Review 47:321–335. O’Rourke, K. H., and R. Sinnott. 2001. ‘‘What Determines Attitudes Towards Protection? Some Cross-country Evidence.’’ In S. M. Collins and D. Rodrik, eds., Brookings Trade Forum 2001, 157–206. Washington, DC: Brookings Institute Press. ———. 2006. ‘‘The Determinants of Individual Attitudes Towards Immigration.’’ European Journal of Political Economy, forthcoming.
138
Heckscher-Ohlin Trade Theory
O’Rourke, K. H., and J. G. Williamson. 1999. Globalization and History: The Evolution of a Nineteenth-Century Atlantic Economy. Cambridge, MA: MIT Press. Rodrik, D. 1997. Has Globalization Gone Too Far? Washington, DC: Institute for International Economics. Roy, A. D. 1951. ‘‘Some Thoughts on the Distribution of Earnings.’’ Oxford Economic Papers 3:135–146. Scheve, K. F., and M. J. Slaughter. 2001a. ‘‘Labour Market Competition and Individual Preferences Over Immigration Policy.’’ Review of Economics and Statistics 83:133–145. ———. 2001b. ‘‘What Determines Individual Trade-Policy Preferences?’’ Journal of International Economics 54:267–292. Tomz, M., J. Wittenberg, and G. King. 1999. CLARIFY: Software for interpreting and presenting statistical results. Version 1.1.1. Available at http://gking.harvard.edu.
Part III
Historical Applications of Heckscher-Ohlin Theory
6
Mediterranean Trade in Biblical Times Peter Temin
This essay analyzes trade in the Mediterranean Sea in biblical times, revealing that the forces Heckscher identified as stimulating trade were effective even before coinage was invented. The first part of the essay describes how we have come to be aware of this trade and what this trade appears to have been like. The second part inquires into the economics of this trade, inferring economic actions and organizations from the physical evidence. Heckscher said in his classic 1919 paper, ‘‘The prerequisites for initiating international trade may thus be summarized as different relative scarcity, that is, different relative prices of the factors of production in the exchanging countries, as well as different proportions between the factors of production in different commodities’’ (Heckscher and Ohlin 1991:48). In the modern world, we reason as Heckscher did from relative scarcities and prices to the existence of international trade. When we study the ancient world, we are at the mercy of our sources, that is, to the accidents of history by which some evidence is preserved and much is lost. We therefore have to reason in the reverse direction, from the existence of trade to relative scarcities and prices. Some of these backward inferences can be verified from the archaeological record, but typically not prices. Archaeology Underwater archaeology until recently was confined to the exploration of shallow waters. There is no lack of ancient shipwrecks in the coastal waters of the Mediterranean, and some of them have been found and described. It seemed reasonable to archaeologists that these ships were representative of ancient shipping, as opposed to representing what
142
Historical Applications of Heckscher-Ohlin Theory
they could find. They therefore inferred that ancient shipping followed the coastline. This view has changed in the past generation as our ability to explore the seabed has improved. The biggest single innovation was in sonar, where ‘‘Doc’’ Edgerton developed side-scan sonar. This new technology allowed modern investigators to find shipwrecks in deep water, even if they were imbedded in the mud bottom. Sonar has been used to find modern wrecks, from the Bismarck in the Atlantic Ocean to Israeli warships in the Mediterranean. It also was used to find ancient shipwrecks in deep water. Deep-water wrecks are very different from those in shallow water. Wrecks in shallow water landed on the bottom at all angles. They also typically have been buffeted by tides and currents, disturbed by fishermen, or looted by earlier explorers. Ships that foundered in deep water sank far through the water to the sea bottom, turning upright as the shape of their hulls offered the least resistance to the water. The ships then sat on the bottom or sank into the mud bottom upright, as if at sea. The wood used in ship construction rotted or was eaten, leaving the cargo in place or laying it down in a kind of projection of the ship onto the sea floor. Two such sunken ships date from the latter half of the eighth century BCE, and they provide the central evidence for this paper. This evidence will be supplemented by customs records a little later and from older coastal wrecks. I need at this point to acknowledge my debt to the marine archaeologists who have found and described these relics of the past, and to Lawrence Stager of Harvard University, who has guided me through the archaeological literature. The two eighth century ships were found far from the shore on a line from the northern Sinai coast and the Nile Delta. Ancient ships had their galleys in the rear, which indicates that the ships probably were going west toward Egypt. (This is only a probable determination since the ships may have been fleeing a storm when they foundered.) The ships might have been destined for the new Phoenician colony at Carthage, but their location and cargo suggest they were bound for Egypt. The ships can be identified as Phoenician and dated to the eighth century by the pottery that was on board. The two ships were 14–14.5 meters long and 5.5–6 meters wide. Their wide berth, one-third of their length, marks them as cargo ships rather than the sleeker and faster warships of the time (Casson
Mediterranean Trade in Biblical Times
143
1991:75–80). They are roughly the same size as a thirteenth century BCE shallow-water wreck. Each of the ships was full of amphoras, approximately four hundred to a ship. Amphoras were the common container of the ancient world, like the plastic gallon jugs of the modern era. Amphoras were made in three parts, a cylinder about as long as a potter’s forearm, a pointed bottom, and a top with a small opening. The shape seems odd to us, living in a world of flat surfaces. But amphoras were made to be packed into ships with sloping sides, in overlapping rows held in place by ropes through their small handles. They were stored on land in structures with dirt or sand floors into which the amphora bottoms could be sunk. They were not used much for overland transport. The amphoras, surprisingly, are of roughly the same size. Eleven of them have been raised, and they average eighteen liters capacity, ranging only from sixteen to twenty liters. Their external dimensions were even more uniform. The presence of four hundred virtually identical amphoras on each of these ships suggests strongly that some form of trade was under way when these ships foundered. A computergenerated image of the shipwreck as it was found, showing the amphoras still in some sort of order on the sea floor, is shown in figure 6.1. This picture was the first evidence I saw of this trade, and it stimulated me to write this essay. These amphoras were made in Phoenicia, an identification that can be made definitively from the presence in the pottery of an algae found only in coastal deposits of what is now Lebanon. They were made in a shape found in sites around that area, a shape that was recognizable elsewhere as being from Phoenicia and quite possibly as a guarantee of the standardized volume of the amphoras. A few amphora ‘‘factories’’ consisting of several kilns have been found from this period, and it is likely that the amphoras in our ships came from one or two of them. Amphoras made in Phoenician style have been found in many Egyptian sites of slightly later dates, suggesting that these ships—or other ships if not these particular ones—could have been destined for Egypt (Maeir 2002). The amphoras originally contained wine, though they were full of mud, not wine, when found. The unbaked stoppers on the amphoras had dissolved, and mud had scoured out the amphora interiors. The mud did not clean the amphoras completely, and we can see that they had been lined with pine pitch from a common conifer found at
144
Historical Applications of Heckscher-Ohlin Theory
Figure 6.1 An eighth-century shipwreck as it ‘‘appears’’ today Source: Ballard et al. (2002:156). Courtesy of H. Singh and J. Howland
the time in the Mediterranean area. This resin lining had trapped tartaric acid, found in grapes and grape products, from the amphoras’ contents. The Phoenician hinterland produced wine, some of which was regarded highly in ancient sources. The wine would have been grown away from the coast and shipped overland to the Phoenician coast. It appears to have been carried in dannu vessels that contained ten times as much as amphoras. Ezekial spoke of daneˆ yayin me’uzal, which can be translated as ‘‘(large) containers of wine from Izalla,’’ an ancient town in modern Turkey (Ballard et al. 2002). Far earlier in time, the discovery of a thirteenth century BCE shipwreck off the coast of Turkey, the Uluburun wreck, ‘‘is yet another indicator of an eastern Mediterranean sea route for the east-west transport of copper, tin, and other raw materials during the Late Bronze Age’’ (Pulak 1998:191). This ship contained about 500 copper ingots, 150 jars containing glass beads, olives, and—mostly—what appears to be resin. The ship may have been transporting wine like the later ships, but this ship may have been simply transporting the resin as there is
Mediterranean Trade in Biblical Times
145
no tartaric acid in these amphoras. There also was a variety of glass beads, jewelry, and ceramics. And there were some weapons: four short swords and several daggers, arrowheads, and spearheads. The shipwreck contained remains of three balances and almost 150 weights. Half the weights had been carefully finished in a variety of zoomorphic shapes. The weights are not marked to indicate their weight, but we presume that ancient merchants did not have to choose among the full set. They each probably had a few sets, whose weights were easily distinguished. Archaeologists, being without the cloth or leather pouches that originally contained the weights, have had to search for order among them. They hypothesized a base weight and looked to see if many weights were simple multiples or fractions of this weight. Assuming a standard unit of 9.3 grams, they found weights that appeared to range from simple fractions of this value up to fifty units. Not all weights fit into this scheme, and statistical analysis suggested that there were at least three different weight standards represented on this ship. The main standard, based on 9.3 grams, ‘‘undoubtedly represents a shekel of the ‘Syrian’ standard, a standard that was based on the Egyptian qedet’’ (Pulak 2000:259). The second most frequently found group of weights was slightly lighter, about 8.3 grams, and corresponded to a Mesopotamian standard. A third group was even lighter, and there was a hint from a few weights of even a fourth group. We have also from about this time a Theban tomb painting that shows the scene of Syrian ships arriving in Egypt. The scene records several ships, some in the process of discharging their cargo, ranging from amphoras to live bullocks. Near the ships are three stalls where Egyptian men and women are engaged in what appear to be sales. They have textile samples hanging with other possible goods below, and two men in separate booths are holding balance scales. The disposition of most of the cargo is unclear, as the surviving picture shows them being carried to a missing destination. Archaeologists therefore have interpreted the stalls as private enterprise set up to deal along the margins of trade. This painting clearly shows the presence of market transactions in interregional trade, but it leaves open the extent to which the trade was bought and sold, as opposed to being accepted as tribute or other obligation (Davies and Faulkner 1947). The Uluburun wreck and Egyptian painting reveal the existence of trade in the Bronze Age. There was an interruption in trade, or at least in our evidence for it, in the transition between the Bronze and Iron
146
Historical Applications of Heckscher-Ohlin Theory
Ages in the twelfth century BCE. Once stability returned, trade appears to have revived in old patterns, as shown by the two eighth-century shipwrecks. The evidence from the wrecks can be augmented by an Aramaic palimpsest on papyrus (that is, papyrus written on more than once with the underwriting still visible) containing records of the inspection, registration and taxation of ships arriving in Egypt for ten months of the fifth-century year, 475 BCE. The largest of the ships contained almost 1,500 amphoras of wine, four times as much as the two eighth-century ships. (It was about the same size as Columbus’s Nin˜a.) Other ships also contained amphoras of wine, typically in larger numbers than in the earlier ships, as well as copper, tin and iron ingots. Phoenican ships as a whole brought over 6,000 amphoras of wine to Egypt in the fall of 475 BCE, indicating a large demand for northern wines in Egypt. This large volume of Egyptian wine imports in the fifth century strengthens the case for an Egyptian destination for the eighth-century ships. Economics The first point to make is that there is abundant evidence of trade in bulk commodities even before the invention of coinage, much less of more sophisticated arrangements. This is powerful support for Heckscher’s argument that regional differences made for trade. Water transportation was far cheaper than land transportation, as it would stay until the invention of the railroad after the Industrial Revolution. The Mediterranean Sea provided a route for long-distance transportation that was cheap enough to use for grain, oil, and wine. Local advantages therefore made for interregional trade. This trade required a lot of coordination. It cannot have been a solitary activity to engage in trade, for the simple matter that there needed to be suppliers and recipients. In addition, not all products were found or grown at seaside; they had to be brought to a port and transshipped to a boat. When they arrived, the goods probably had to be transported again to the location of consumption. As we have seen, containers were needed for liquid cargoes—and probably for solid ones as well. These containers had to be manufactured and allocated to trade. The extent of this coordination can be surmised from the presence of hundreds of virtually identical amphoras in each ship. There must have been many groups involved in this trade: growers, land transporters, transshippers, amphora makers, ship operators, receivers, and
Mediterranean Trade in Biblical Times
147
consumers. There may have been middlemen in addition at several points. Coordination of all these far-flung and quite disparate people was a formidable task. How was this coordination achieved? There are only a few models in the literature of ways to organize such complex interactions. Pryor (1977) provided a useful taxonomy. He distinguished between what he called exchanges and transfers. Exchanges are balanced transactions where goods or services are exchanged for other goods or services of equal value. This of course is the kind of behavior most often observed in markets. Transfers are one-way transactions where goods and services are given without a direct return. Grants, tributes, and taxes are all transfers. Pryor excluded ‘‘invisibles’’ from this accounting, so that taxes are considered to be transfers rather than an exchange of goods or money in order to purchase social order or military success.1 Pryor subdivided exchanges into those in which the ratio of goods or services exchanged can vary and those in which it cannot. The former may or may not involve money; the latter do not. He termed the former ‘‘market exchange,’’ the latter ‘‘reciprocal exchange.’’ The use of money is a good index of this distinction, as are changes in the exchange ratio over time. In the presence of money, of course, changes in exchange ratios are expressed as changes in prices. Pryor divided transfers into centric and noncentric ones. Centric transfers are between individuals in a society and ‘‘an institution or an individual carrying out a societal-wide role’’ (Pryor 1977:34). Heckscher, of course, analyzed market exchanges, and we need to ask if the ships we have found were engaged in this kind of reciprocal activity. The evidence from the eighth century is ambiguous; the ships may have been going to Phoenician colonies or to Egypt. If the former, we have no way of knowing if they were transfers or exchanges. The presence of balances aboard the Uluburun wreck and in the Egyptian painting, however, indicates the prevalence of market activity. The key as always is how these artifacts were used. What were the ancient traders doing with their balances? The archaeological evidence indicates that they were comparing the weights of materials to some standard. In other words, they were dividing whatever was being weighed into units of a standard, identified by Pulak as a shekel. We normally speak of a standard of value as money, although we ask that any putative money have more than this single attribute. The other functions of money are as a means of transaction and a store of value. There is no indication that the weights
148
Historical Applications of Heckscher-Ohlin Theory
found in the Uluburun shipwreck were traded, and we must look elsewhere for money performing these other functions. Related evidence suggests that silver was used for transactions and as a store of value, divided into shekel units. There were no coins as yet, and it was the weight of silver that determined the value of a transaction. Numerous written examples point to the use of silver for transactions. We do not find loose silver, perhaps because it dissipates in the water, but we do find what is called Hacksilber: cut-up silver jewelry, ingots, or figurines that appear to have been stored in fabric pouches. These jewelry fragments could not have functioned as coinage, but they easily could have functioned as silver in larger units than loose silver. If so, they probably were both means of exchange and stores of value. While we cannot infer that all trades across the Mediterranean in biblical times were exchanges, we can confidently assert that some of it—perhaps most of it—was market exchange. Heckscher phrased his proposition in terms of relative prices. We do not observe prices until the end of this period, but there was knowledge of relative scarcities even if people did not convert them into prices. And when we do have records that appear to be prices at the end of this period, they have the characteristics of modern prices. For example, Slotsky (1997) recorded what appeared to be a series of monthly market prices for six agricultural commodities for four hundred years in ancient Babylon, starting in the fifth century BCE. They appeared to provide much more evidence of ancient market activity than had been available earlier, and Slotsky argued that her observations were market prices. I conducted an econometric analysis of these prices and confirmed that they indeed were market prices; they moved with a great deal of randomness, and they varied over time. More precisely, the Babylonian agricultural prices moved like the random walk of modern prices, and they varied together in response to exogenous events that affected all crops. These changes are clearly understood within a market framework; they are impossible to understand within an administrative one. I concluded therefore that the scribes recorded prices set in functioning markets, that is, they were examples of what Pryor called market exchanges (Temin 2002). I infer from this mixture of evidence from different times and places that much of the Mediterranean trade of the biblical era was exchange, coordinated by market activities; it was market exchange. If so, then the trade was in response to the forces analyzed by Heckscher, a re-
Mediterranean Trade in Biblical Times
149
sponse to differing comparative advantage around the Mediterranean basin. Given the technology of the time, wine could not be produced in North Africa. It was imported from more northern localities. Egypt exported wheat as shown by the description of the traders of Sidon in Phoenicia in Isaiah (23:2–3); they imported grain from Shihor (Lower Egypt), which was ‘‘the harvest of the Nile.’’ It had been the reliable result of the annual flooding of the Nile since the times of Joseph and his brothers and would remain so into Roman times. Egypt also exported natron, a form of sodium carbonate used to make glass, bleach textiles, and treat sick people, that was found in the Nile delta, but not on the northern shores of the Mediterranean. Even if these products were not all the object of market exchanges, they all were articles of reciprocal trade. Heckscher talked of relative factor scarcities as well as relative scarcity of goods. It is even harder to find evidence of returns to factors than of product prices in this era, but we can at least pose the question. It is likely that land was the scarce factor in Egypt, since the agricultural area was limited to the flooding of the Nile. If trade was the result of differing factor proportions rather than different climates, if must have been the case that wheat was a labor-intensive commodity, while wine was a land-intensive one. In Phoenicia, trade raised the relative price of wine and therefore, if the assumption of relative factor proportions is correct, the return to land. The opposite effect in Egypt may well not have been relevant since Egypt was specialized in wheat production. Phoenician landowners may have been the primary gainers from trade as well as its apparent instigators. These gains, however, were offset in part by increased risks. Three kinds of risks can be identified. There was the risk of natural disaster along the way, of the ship sinking or getting lost. There also was a risk of trouble in the form of attack by pirates along the way or hostile people when the ship landed. And there was the risk of not being paid, or not paid what was due, by the recipients of the cargo. I discuss them in turn. While it certainly was bad for ancient sailors and traders that sailing was hazardous in the ancient world, we are the beneficiaries. If we did not find ships at the bottom of the sea, we would have only a few clues to the existence of ancient trade. We cannot know if the risk changed over time. Archaeologists and ancient historians have tended to assume that the risk of shipwreck stayed constant; they have used the frequency of wrecks as an index of the volume of trade (Hopkins 1980).
150
Historical Applications of Heckscher-Ohlin Theory
Ancient mariners were conscious of these risks, and they took actions to minimize them. They did not sail, or not often, in the winter. The port records in Egypt showed that ships vanished only in January and February, a short winter even for the Mediterranean. Roman shipping at a later date stopped for a longer period in the winter (DuncanJones 1990, chapter 1). It was thought earlier that sailors also reduced the risk of loss by staying close to shore. We now know that some ship captains were willing to brave the open seas. Given the primitive tools for navigation available at the time, they probably were not willing to make voyages for very long out of the sight of land. Hence the importance of trade along the hypotenuse of a triangle at the corner of the Mediterranean or across the strait opposite Carthage. These probably were the deep-water shipping lanes of the biblical era (Ballard et al. 2000). The risk of capture was quite different, but even harder to recover at this late date. A time when governments had limited jurisdiction and no single empire ruled the Mediterranean must have been a time when pirates abounded. Yet the ships that we have found in deep water do not appear to have been fortified. There are three possibilities: either the arms were so small as to be invisible in the wreckage, or they simply have been lost, or there were no pirates to fear. There were some weapons aboard the Uluburun wreck, but they do not seem enough to repel pirates. There also is evidence of warships, leaner and faster than the merchant ships I have described, that would provide another defense against pirates. Two ships containing eight hundred amphoras of wine between them must have been very tempting to a potential pirate. Or was it? Casson (1991:178) asserted that ancient pirates sought people rather than goods. It may have been hard for a pirate to sell eight hundred amphoras of Phoenician wine without the guarantee of both quantity and quality that came from Phoenician merchants. However poor people could be sold as slaves; rich people, held for ransom. The incentives can be illustrated by a story from the years just before the Romans cleared the Mediterranean of pirates. Pirates captured the young Julius Caesar on a voyage in the eastern Mediterranean. They sought and received a large ransom, made larger by Caesar’s boast that he was worth more than their original request. However Caesar, once free, hired ships and soldiers and returned to capture the pirates. He recovered his ransom and crucified the pirates, as he had threatened while still their captive (Plutarch, Caesar, 1–2).
Mediterranean Trade in Biblical Times
151
This story suggests a small theory of ancient piracy. Pirates wanted to find rich and powerful people at risk, because they could earn more from ransom than from the sale of captives as slaves. Human capital was worth more to the relatives and colleagues of captives than to strangers who might purchase slaves. This relation was not monotonic, however. Capturing someone who was exceedingly rich or powerful was potentially dangerous as well as profitable, as the story about Caesar illustrates. There must have been an optimum wealth for ancient pirates to aim for, quite possibly one that shifted over time. The existence of pirates in earlier times can be inferred from the political record. The break between trade in the Bronze and Iron ages noted earlier was due to the invasion of the ‘‘Sea Peoples.’’ This invasion appears in the literature as a political operation, but it had an important economic dimension as well. Much later, Norsemen invaded England and France in the ninth century CE, starting with piracy and marauding and progressing to settlement (Bloch 1961). The earlier invasion, about which we know far less, undoubtedly exhibited the same progression. The heightened risk of piracy provides a good explanation for the dearth of Mediterranean trade in the twelfth century BCE (Sherratt and Sherratt 1993:366). Assuming the ship arrived safely to its intended port, there also was the risk of confiscation or commercial double-dealing. There were no international treaties or World Trade Organization to monitor these transactions. There must have been less formal arrangements that shippers could count on enough to launch two ships containing four hundred amphoras of wine each into the Mediterranean. The most obvious of these arrangements is a reputation equilibrium, that is, an arrangement where people receiving shipments deal fairly with importers and are known to do so. They maintain this behavior in order to encourage trade, forgoing the obvious short-term gain from confiscating a new cargo. This organization was used in Bronze Age Assyrian overland trade, where family firms prospered by sending out sons and other relatives to make the actual sales (Larsen 1976:92–105). Almost three millennia later, the Maghrebi traders of Alexandria had a highly developed form of reputation equilibrium. They sent associates, typically family members and almost exclusively fellow Jews, around the Mediterranean to conduct repeated transactions. They expected their agents to deal honestly with them, but they did not rely solely on this expectation to do business. If an agent cheated, the injured trader shared this
152
Historical Applications of Heckscher-Ohlin Theory
information widely with his colleagues. The resulting loss of reputation was reflected immediately in a loss of employment. An agent who cheated then had the prospect of losing not only his present employment, which he would have lost when he cheated in any case, but also any future employment as a trading agent with the Maghrebi traders. The agent would not cheat under any reasonable circumstance, since the anticipated cost of doing so would be far larger than the gain he could get from pocketing the results of a single voyage (Greif 1994).2 These motives were in force also in the early Iron Age, but they were not totally compelling. The Israelite prophets noted the use of false balances and dishonest scales among the other evils of their contemporaries (Amos 8:6; Hosea 12:8). In modern parlance, the transaction costs of international trade in the biblical era were very high. We presume that the gains from trade were even higher, which is why traders found it worthwhile to ship even bulk commodities like wine and wheat across the sea in normal times. After trade revived in the Iron Age, in the eighth century, traders and Israelite kings appear to have made efforts to reduce transaction costs in trade. They appear to have done this in a variety of ways. Traders began to label their weights with the units in shekelim that the weights represented. This enabled their transactions to be monitored more easily. The Israelite kings introduced jars labeled lemelek, meaning ‘‘of the king,’’ which probably functioned as guarantors of the standardized volume of the jars. Of course, if the king owned the contents of the jars as well as the jars themselves, then the thousands of lemelek inscriptions indicate centric transfers rather than market exchanges. The reforms of Hezekiah (eighth century) and Josiah (seventh century), which have been celebrated for their religious aspect, appear also to have contained attempts, which evidently had to be repeated, to standardize weights and measures (King and Stager 2002:312–314). All of these actions must have reduced transaction costs by increasing standardization and verifiability. They encouraged international trade and market exchanges—if they did not indicate the growth of a centralized state with centric transfers. Prevailing archaeological thought supports the existence of markets rather than a centralized state. Larsen (1987:54) described ‘‘a highly interactive world in which it is possible to follow commodities flowing from one end to the other’’ in the Middle Bronze Age. The Bronze Age trade was organized as a mixture of controlled flows directed by central authorities and commercial transactions, but it was primarily
Mediterranean Trade in Biblical Times
153
limited to overland trade east of the Mediterranean. Liverani (1987) described the shift from the Bronze Age to the Iron Age as a shift from centralized empires to trading city-states, from a pattern of gift exchanges to profit-oriented commercial activity. (In Pryor’s terminology, the shift was from reciprocal to market exchange.) Sherratt and Sherratt (1993:362) agreed, stating that, ‘‘merchant enterprise, rather than state-controlled exchange, became the dominant mode of trading activity’’ of Mediterranean trade in biblical times. Archaeologists, however, have focused primarily on trade in luxury goods, perhaps because they find evidence of them in archaeological sites. The point I want to make here is that transport and transaction costs fell enough by the Iron Age to create incentives for extensive international trade in bulk commodities. A vivid window into transaction costs is given by the account of Wen-Amon’s trip from Egypt to Phoenicia around 1050 BCE. This was about fifty years after the troubled time when the ‘‘Sea Peoples’’ destroyed the established networks of trade; the ‘‘Sea Peoples’’ in the Levant had changed from raiders to traders. Wen-Amon tells the story in the first person, and we presume that he was able to complete his journey, even though the surviving narrative is incomplete. We do not know if he wrote this account because it was typical or atypical; in the absence of other accounts, we take it as the former. Wen-Amon went to Phoenicia to buy cedars of Lebanon for use in the Temple of Amon at Karnak. We know Wen-Amon went to buy the cedars because he brought silver with him to pay for them. And when he said to his host that his host should do as his father and grandfather had done in supplying cedars, his host responded that he would do so—as long as Wen-Amon gave him something in return. This was a market exchange, not a centric transfer. The story’s context suggests strongly that the seller expected silver. The historical context was the breakdown of the Egyptian state. But the sale was not completed quickly. Wen-Amon took with him about a kilogram of silver and some gold. However, all this money was stolen as soon as he reached Dor, a port on the eastern shore of the Mediterranean occupied by one of the ‘‘Sea Peoples’’ called the Sikkel, located between the Phoenicians to the north and the Philistines to the south. The Sikkel had been pirates before they were traders, illustrating the transition described earlier (Stager 1995:337). It therefore should not be surprising that Wen-Amon’s appeal for restitution was denied. Archeologists recently have unearthed a jar containing
154
Historical Applications of Heckscher-Ohlin Theory
seventeen linen bags of silver, each weighing half a kilogram, in Dor from about the time of Wen-Amon’s voyage (Stern 1998). Could this jar have been the eventual destination of his silver? Finding himself in a tough place, Wen-Amon stole a comparable amount of silver from another ship and, apparently, used it in place of his own. Negotiations for the cedars dragged on for a long time with lots of histrionic speeches. It took months if not years to make the purchase and arrange for the cedars to be shipped to Egypt. When WenAmon started back to Egypt, he was shipwrecked in a storm in hostile territory. He persuaded his captors not to kill him, and the narrative breaks off (Pritchard 1955:25–29). Three lessons emerge from this colorful tale of commerce in difficult times. The first is how hard it was to conduct international trade when the world was composed of many small political entities. The second is that international trade in quite heavy commodities took place despite these large transaction costs. And the third is that the reforms of the eighth and seventh centuries are easily understood as attempts to reduce these enormous transaction costs. Conclusion This essay bridges two disciplines. I hope that it indicates how economics can inform archaeology, framing questions and focusing the search for evidence. More relevant for this conference, it shows that the forces leading to international trade are very old. We know that trade in luxury goods and special items existed since time immemorial; the added information here is that trade in bulk commodities was present in the early Iron Age if not even earlier. I have argued here that this trade conforms to the patterns analyzed by Heckscher that have become staple items in the analysis of more recent trade. One does not need to have modern ships or communication technology or even coinage to engage in extensive international trade. It is unlikely that there was anything like factor price equalization in the Iron Age, but I have argued that there were tendencies in that direction. Acknowledgments I thank Lawrence E. Stager for his extensive advice, encouragement, and help and M. June Flanders and Elhanan Helpman for their helpful comments and suggestions. All errors, of course, are mine alone.
Mediterranean Trade in Biblical Times
155
Notes 1. This exclusion is necessary because one can always hypothesize an invisible gain that makes all transactions balanced. In that case, there is no way to discriminate between different forms of behavior. 2. Larsen referred to the medieval traders in explaining his views on Assyrian trade, albeit to authors who predated Greif.
References Ballard, R. D., L. E. Stager, et al. 2002. ‘‘Iron Age Shipwrecks in Deep Water off Ashkelon, Israel.’’ American Journal of Archaeology 106:151–168. Ballard, R. D., A. M. McCann, et al. 2000. ‘‘The Discovery of Ancient History in the Deep Sea Using Advanced Deep Submergence Technology.’’ Deep-Sea Research 47:1591–1620. Bloch, M. 1961. Feudal Society. London: Routledge and Kegan Paul. Casson, Lionel. 1991. The Ancient Mariners. Princeton, NJ: Princeton University Press. Davies, N. de G., and R. O. Faulkner. 1947. ‘‘A Syrian Trading Venture to Egypt.’’ Journal of Egyptian Archaeology 33:40–46. Duncan-Jones, R. 1990. Structure and Scale in the Roman Economy. Cambridge: Cambridge University Press. Greif, A. 1994. ‘‘Cultural Beliefs and the Organization of Society: A Historical and Theoretical Reflection on Collectivist and Individualist Societies.’’ Journal of Political Economy 102:912–950. Heckscher, E. F., and B. Ohlin. 1991. Heckscher-Ohlin Trade Theory. Cambridge, MA: MIT Press. Hopkins, K. 1980. ‘‘Taxes and Trade in the Roman Empire (200 B.C.–A.D. 400).’’ Journal of Roman Studies 70:101–125. King, P. J., and L. E. Stager. 2002. Life in Biblical Israel. Louisville, KY: Westminster and John Knox. Larsen, M. T. 1976. The Old Assyrian City-State and Its Colonies. Copenhagen: Akademisk Forlag. ———. 1987. ‘‘Commercial Networks in the Ancient Near East.’’ In M. Rowlands, M. Larsen, and K. Kristiansen, eds., Centre and Periphery in the Ancient World, 47–65. Cambridge: Cambridge University Press. Liverani, M. 1987. ‘‘The Collapse of the Near Eastern Regional System at the End of the Bronze Age: The Case of Syria.’’ In M. Rowlands, M. Larsen, and K. Kristiansen, eds., Centre and Periphery in the Ancient World, 66–73. Cambridge: Cambridge University Press. Maeir, A. M. 2002. ‘‘The Relations Between Egypt and the Southern Levant During the Late Iron Age: The Material Evidence from Egypt.’’ A¨gypten und Levante 12:235–246. Pritchard, J. B. 1955. Ancient Near East Texts Relating to the Old Testament. 2nd ed. Princeton, NJ: Princeton University Press.
156
Historical Applications of Heckscher-Ohlin Theory
Pulak, C. 1998. ‘‘The Uluburun Shipwreck: An Overview.’’ International Journal of Nautical Archaeology 27:188–224. ———. 2000. ‘‘The Balance Weights from the Late Bronze Age Shipwreck at Uluburun.’’ In C. F. E. Pare, ed., Metals Make the World Go Round: The Supply and Circulation of Metals in Bronze Age Europe, 247–266. Oxford: Oxbow. Pryor, F. L. 1977. The Origins of the Economy: A Comparative Study of Distribution in Primitive and Peasant Economies. New York: Academic Press. Sherratt, S., and A. Sherratt. 1993. ‘‘The Growth of the Mediterranean Economy in the Early First Millennium BC.’’ World Archaeology 24:361–378. Slotsky, A. L. 1997. The Bourse of Babylon: Market Quotations in the Astronomical Diaries of Babylonia. Bethesda, MD: CDL Press. Stager, L. E. 1995. ‘‘The Impact of the Sea Peoples (1185–1050 BCE).’’ In T. E. Levy, ed., The Archaeology of Society in the Holy Land, chapter 20. London: Leicester University Press. Stern, E. 1998. ‘‘Buried Treasure: The Silver Hoard from Dor.’’ Biblical Archaeology Review 24:46–62. Temin, P. 2002. ‘‘Price Behavior in Ancient Babylon.’’ Explorations in Economic History 39:46–60.
7
Demographic Shocks and the Factor Proportions Model: From the Plague of Justinian to the Black Death Ronald Findlay and Mats Lundahl
All of his professional life, Eli Heckscher was concerned with the methodology of economics and economic history. Straddling both disciplines, it was essential for him to come to grips with the problem of what one discipline could learn from the other and vice versa (Findlay 1998; Findlay and Lundahl 2002; Henriksson and Lundahl 2003). Gradually he also introduced the use of statistical time series in his works, notably his four-volume magnum opus about the economic history of Sweden from the time of Gustav Vasa (1523–1560) to ‘‘the present,’’ which in practice meant the early nineteenth century. As he moved from the sixteenth to the eighteenth century the availability of quantitative material increased, and Heckscher made use of it. However, he steadfastly refused to be bound by the strict limits imposed by the ‘‘hard’’ facts when it came to the interpretation of a certain epoch. He certainly took great care to weed out hypotheses not grounded in facts, but he was no stranger to hypothetical reasoning either. Such reasoning was needed for arriving at a historical synthesis, and the refusal to go beyond just what the sources would reveal, after thorough critical scrutiny, was for him to stop short of attempting a synthesis. The present essay should be seen as the effort of two economists to provide a building block for a historical synthesis. We would like to combine Heckscher’s plea for economic theory in economic history with his insistence on historical synthesis into a plea for economic theory in historical synthesis. The roots of the essay are found in economic theory, which we employ very much in the fashion that Heckscher used it in his historical works: as a device for explaining the equilibrating processes in economic history during a determined period. We also share with Heckscher the conviction that it is difficult to use economic theory to explain transitions from one historical epoch or era to another. Here factors exogenous from the point of view of
158
Historical Applications of Heckscher-Ohlin Theory
economics must be invoked, and we begin and end our story with two such events: the so-called Plague of Justinian in the mid-sixth century AD and the Black Death eight hundred years later. What our analysis deals with is precisely the period between these two events, and the synthesis that concerns us is one based on a marriage between the historical facts, as provided by the professional historians, and an economic theory that may serve as reading glasses when it comes to the interpretation of the events. Our aim is to provide a reading of some of the fundamentals of eight centuries of the history of Western Europe, Eastern Europe, and Islam that is consistent with the interplay of the probable basic economic mechanisms at work; that is, we want to put history on a sound theoretical footing, to heed Heckscher’s (1929) plea for the use of theory in economic history, and simultaneously to allow economic history to influence theorizing. Heckscher made it clear that theory was useful when it came to organizing the questions to be put to the historical material, and we agree, but any old theory won’t do. In order to be efficient, our theory must spring out of the historical material itself. Its categories must be chosen in a way that properly reflects the characteristics of the period we have chosen to deal with. It is only through the interaction between the given historical material and the specific tools we need to construct to reveal its hidden secrets that we can arrive at a better understanding of the underlying processes that made economic history take a particular turn at given moment in time. The Plague of Justinian and Its Aftermath The waves of bubonic plague that swept the world of Late Antiquity and the Early Middle Ages have taken their name from the Emperor Justinian I, ruler of the eastern Roman Empire from 527 to 565. It was during his reign that the first outbreak occurred in Constantinople in 542, killing hundreds of thousands of people according to the historian Procopius, secretary to Justinian’s general Belisarius. Procopius, who was an eyewitness, has provided us with the first incontestable description of bubonic plague in history (quoted by Bray 2000:22–23): During these times there was a pestilence, by which the whole human race came near to being annihilated, . . . it did not come in a part of the world, nor upset certain men, nor did it confine itself to any season of the year . . . but it embraced the whole of the world, and blighted the lives of all men, though differing one from another in he most marked degree, respecting neither sex nor age . . .
Demographic Shocks and the Factor Proportions Model
159
With the majority it came about that they were seized by the disease without becoming aware of what was coming. They had a certain fever. . . . The body showed no change from its previous colour, nor was it hot . . . nor did any inflammation set in. . . . But on the same day, in some cases, in others on the following day, and in the rest not many days after, a bubonic swelling developed; and this took place not only in the particular part of the body which is called the ‘‘boubon,’’ that is below the abdomen, but also inside the armpit, and in some cases also beside the ears and at different points on the thigh . . . . For there ensued with some a deep coma, with others a violent delirium . . . for neither physicians nor other persons were found to contract this malady through contact with the sick or with the dead, for many were constantly engaged in burying or attending those in no way connected with them. . . . With some the body broke out in black pustules about as large as a lentil and these did not survive even one day, but all succumbed immediately. With many also a vomiting of blood ensued without visible cause and straightaway brought death. . . . In some cases when the swelling rose to an unusual size and discharge of pus had set in, it came about that they escaped from the disease and survived.
The emperor himself was infected, but survived. The first wave of the 540s–550s was followed by several others of varying intensity, persisting intermittently until the middle of the eighth century (Biraben and LeGoff 1969). While there is wide disagreement as to the extent of the impact (cf., e.g., Harrison 1999:141–153, for a critical view), there is little doubt that it was a demographic catastrophe on a scale not exceeded till the Black Death of the fourteenth century. Russell (1968:180) estimates a 20–25 percent loss for the first epidemic of 541– 544 and a total loss of 40–50 percent of the pre-plague population over the period 540–700. The plague is believed to have entered the Mediterranean world at the Egyptian port of Pelusium through Ethiopia and the Red Sea, before spreading both east and west. The populations of Egypt, Syria, and Palestine, then all under the sway of the Byzantine Empire, were severely affected. What happened farther east is not clear (Bray 2000:27): Its distribution to the east of Syria is something of a puzzle. It would seem that the Arabs did not bring the plague back to the Hejaz. . . . Equally it would seem that the Arabs did not bring he plague to Afghanistan, the Indus, Ferghana and Transcaucasia and thus into the Indian subcontinent and China, as all observers have the plague ceasing at about the present western border of Iran and at the Caucasus. Whether it penetrated Africa south of Egypt is anyone’s guess.
The Middle East as well was hit by successive outbreaks of the plague. Dols (1974) lists six ‘‘major’’ epidemics between 627 and 717.
160
Historical Applications of Heckscher-Ohlin Theory
The westward spread to Italy, Gaul, and Spain was through the seaports of Genoa, Marseilles, and Narbonne, proceeding inland along the rivers and trade routes. The impact on the northern lands seems to have been much less severe, though not necessarily negligible. The arrival of the plague at Marseilles in 588 was vividly described as follows by the contemporary Gallo-Roman historian Gregory of Tours (1974:510–511): A ship from Spain put into port with the usual kind of cargo, unfortunately also bringing with it the source of the infection. Quite a few of the townsfolk purchased objects from the cargo and in less than no time a house in which eight people lived was left completely deserted, all the inhabitants having caught the disease. The infection did not spread through the residential quarter immediately. Some time passed, and then, like a cornfield set alight, the entire town was suddenly set ablaze with the pestilence . . . at the end of two months the plague burned itself out. The population returned to Marseilles, thinking themselves safe. Then the disease started again and all who had come back died. On several occasions later on Marseilles suffered from an epidemic of this sort.
Before the plague struck, the eastern Roman Empire under Justinian was at the height of its power, with plentiful resources of manpower and revenue. Not only was he able to secure the eastern frontier against the Persians but also to stabilize and push back the frontiers against the Avars, Lombards, Berbers, and others, while soundly defeating the Vandals in North Africa in 534–535. The great scheme of once again unifying the eastern and western halves of the empire was not impossible, though of course extremely difficult. The devastating effects of the plague on both manpower and revenue, however, rendered it impossible. Nevertheless, although his plans for expansion failed he was still able to retain most of the eastern territories of the empire. As often noted, the nomads of the Arabian Peninsula escaped the ravages that the plague wrought on the more settled Byzantine and Sassanid empires. After the unification of the tribes under the Prophet and his early successors, the ‘‘rightly guided’’ Caliphs, at the beginning of the seventh century, the Arabs rapidly captured Syria, Palestine, Egypt, North Africa, and western Mesopotamia from the Byzantines. Sassanid Iran was conquered later by the Arabs under the Ummayad Caliphate. The Byzantine emperor Heraclius (610–641) and his successors could not match the e´lan of the Arab onslaught with the depleted resources at their command. They retreated to the Anatolian Plateau,
Demographic Shocks and the Factor Proportions Model
161
behind the security of the Taurus mountain range, and then successfully resisted successive Arab attacks on Constantinople from the sea. The empire was to survive for eight hundred more years but it had lost the bulk of its territory, population and revenue to the Arabs. In the west, the empire’s hold on Italy was rendered extremely tenuous, and the Lombard invasion of the peninsula, beginning in 568, could not be checked before most of it had been occupied. Gaul was left to the Merovingian Franks. Spain, under Visigothic rulers, was invaded in by the Muslims in 711, who established the powerful Emirate of Cordoba that occupied most of the Iberian Peninsula. The popes in Rome ceased to look to Constantinople and eventually allied themselves with the rising dynasty of the Frankish warlord Charles Martel, who checked the Arab advance at Poitiers in 732 or 733.1 Around the same time, say 718 or 722 (Collins 1995:182), after a Christian victory over Muslim forces at Covadonga in Asturias, the Spanish Reconquista began. By the second half of the eighth century the territory of the Roman Empire at its height had come to be divided between three great powers, the Greek Orthodox Byzantine Empire under the vigorous new Isaurian Dynasty, with its capital at Constantinople, the Muslim Abbassid Empire, with its capital at Baghdad and the Latin Catholic Carolingian Empire with its capital at Aachen. Beyond the frontiers of these empires were Anglo-Saxon England in the west, the pagan Scandinavian and Saxon tribes in the north and in the east the Slavs and Bulgars in the Balkans. Population figures for such early periods in history are notoriously unreliable and can only be taken as the best guesses made by scholars on the basis of extremely scanty evidence. Issawi (1981), citing Russell (1968), gives figures of 16.6 million in 350 for the eastern Roman Empire, falling to 10 million by 600, indicating the devastating effect of the early waves of the Plague of Justinian. Treadgold (1997:278) gives a figure of as much as 26 million for the empire under Justinian in 540, before the plague, falling to 17 million in 610 before the Arab invasions and collapsing to 7 million in 780 after that disaster, rising back to 12 million by 1025. Issawi cites 15 million at least for 1000, or about 25 percent more than Treadgold. The general qualitative picture, however, is similar, with a heavy loss of population due to plague, more loss due to the Arab invasions, and then recovery within the restricted territory.
162
Historical Applications of Heckscher-Ohlin Theory
Issawi’s main contribution is an estimate of the area and population of the Arab Empire at its height in 750. He puts the population at between 28 and 36.5 million over the vast arc stretching from Spain to India. The breakdown is 10–13 million in the former Byzantine territories of Egypt, Syria and North Africa; 5–6 million in Spain; 1.5–2 million in the Arabian peninsula; 5–6 million in Iraq; 3–4 million in Iran; and 3.5–5.5 million in Central Asia, Afghanistan, and India. The gross area he puts at 9.8 million square kilometers and the inhabited area at 2.1 million, only slightly more than one-fifth, indicated the highly arid inhospitable nature of much of the terrain. For the tenth to eleventh centuries Issawi estimates only a very modest rise of the population to 35–40 million, or about 25 percent over three centuries. The Abbassid Empire was about twice as big in both population and inhabited area as the contemporary Byzantine Empire. The only comparable source of population history for the world as a whole is McEvedy and Jones (1980). We have checked Issawi’s figures against the comparable estimates that they provide. While their figures are generally lower they are not too far below the bottom range of Issawi. The figures based on their estimates of individual countries are 21 million for 700, 26 million for 1000, and 22 million for 1300. Thus the trend of population for the Islamic world is a modest 25 percent over three centuries for both Issawi and McEvedy and Jones. The fall from 26 million in 1000 to 22 million in 1300 is attributable to the ravages of the Mongol invasion in Iran and Iraq in the east and the Bedouin raids in North Africa in the west. For Europe excluding Russia, McEvedy and Jones report a peak of 33 million in 200, falling to a trough of 23 million in 600, rising back to 32 million by 1000, and then accelerating upward to the huge figure of 70 million in 1300 on the eve of the Black Death.2 This gives annual growth rates of less than 0.1 percent per annum for the earlier period and 0.25–0.3 percent after the turn of the millennium. David Grigg (1980:53) reports some figures for individual countries: 0.43 percent per annum for England and Wales from 1086 to 1340, 0.49 percent for France from 1100 to 1328, less than 0.2 percent for Italy from 950 to 1300 and for Denmark from 1000 to 1300, 0.46 percent for the Moselle valley between 1000 and 1237, and an average rate of increase for Europe between 1000 and 1340 of 0.26 percent. While these figures are notoriously uncertain, representing a tendency only, the beginnings of the higher demographic growth that was to last for three centuries can be dated. ‘‘It was 930–50 in Sabina and Lombardy, 940–90 in Catalonia, 980–1010 in Languedoc, Pro-
Demographic Shocks and the Factor Proportions Model
163
vence, Poitou and the Auvergne, 1010–30 in Flanders and Picardy, Bavaria and Franconia, Burgundy and Normandy, 1050–80 in England and the Rhineland, after 1100 in central Germany’’ (Fossier 1999:62). If we are to believe the McEvedy-Jones estimates, the increase from the trough in 600 to the peak in 1300 is a remarkable threefold or 200 percent over the seven centuries. For comparison, China increased from the 200 peak of 60 million back to 60 million in 1000, just like Europe, before rising further to a peak of 115 million in 1200, before falling to 85 million in 1300 as a result of the Mongol invasions at the other end of the European peninsula. This approximate doubling took place under the Sung Dynasty, generally regarded as the golden age of imperial China. Yet it does not come close to the European achievement. Why did the Western Europe of the so-called Dark Ages perform so spectacularly well, relative to what were at that time the much more advanced Byzantine, Islamic, and Sung Chinese civilizations? Clearly, the foregoing leads us to suspect that the demographic change caused by the plague had something to do with it.3 As it seems, the economic trend of late antiquity was a downward one in the Mediterranean area. In his monumental work on the ‘‘dark’’ centuries of the European economy, Michael McCormick (2001:41) concludes his chapter of the end of the ancient world thus: Settlement patterns suggest that ill-understood processes of demographic stagnation and decline moved slowly across the old Roman space. They finally reached the east in the sixth or early seventh century. Around the same time, the history of disease marks a new configuration in the health experience of the population. The new pathocoenosis [the array of diseases characteristic of a society] could not by itself have been the leading cause of wide-reaching economic change, for change had started earlier in most of the Mediterranean world. But it surely must have reinforced some aspects of that change, if only by debilitating or destroying part of the work force. Both settlement patterns and disease encourage us to believe that the overall trend of the late Roman world was downward between c. 200 and 700.
The next section of this essay will develop a model of demographiceconomic-ecological interaction that could perhaps provide some clues to the very fundamental problem of the unequal patterns of decline and, later, rebirth. The Model The model presented in this section draws on our previous work (Findlay 1993; Findlay and Lundahl 2002) combining the Malthusian
164
Historical Applications of Heckscher-Ohlin Theory
demographic specification with the concept of an endogenous land frontier, while stripping away the other features of those models. The Production Function The economy is considered as producing a single input, denoted Y, with land, A, and labor, L, as inputs, according to the production function Y ¼ YðA; LÞ
ð1Þ
which is taken to have the usual neoclassical properties of constant returns to scale, positive first and negative second derivatives with respect to each input and complementarity between the inputs. Constancy of returns to scale enables us to write y ¼ yðaÞ
ð2Þ
where y and a denote Y and A divided by L, respectively. We can also write the rent per acre and the real wage as r ¼ y 0 ðaÞ
ð3Þ
w ¼ yðaÞ y 0 ðaÞa
ð4Þ
The Malthusian Mechanism The fertility and mortality rates of the population are specified as positive and negative functions, respectively, of the per capita consumption, c, of the population f ¼ f ðcÞ;
f 0 ðcÞ > 0
ð5Þ
m ¼ mðcÞ;
m 0 ðcÞ < 0
ð6Þ
These relations are depicted in figure 7.1 below. The population is in equilibrium at the per capita consumption level of c that equates f ðc Þ and mðc Þ at f and m . Labor Productivity and Per Capita Consumption Labor productivity, as defined above, is y ¼ Y=L
ð7Þ
Demographic Shocks and the Factor Proportions Model
165
Figure 7.1 The Malthusian mechanism
Suppose that we are in a ‘‘stationary state’’ with zero net investment. Total consumption will then be equal to total output and the per capita consumption of the population as a whole will be c ¼ Y=P
ð8Þ
where P denotes the entire population. Suppose that the labor force is proportional to the population so that L ¼ aP
ð9Þ
where a is some fraction. In the preindustrial economic conditions that we are considering, the non-working population will consist of warriors, priests, and other ‘‘unproductive’’ occupations as well as those too old or too young to work. Taking a more ‘‘physiocratic’’ view, we could even include in the ð1 aÞP artisans, traders, and so on who are ‘‘supported’’ by the surplus generated by agriculture, the only truly ‘‘productive’’ sector from this standpoint. From equations (7) to (9) it follows that y ¼ c =a
ð10Þ
so that y is the level of labor productivity that will maintain output, population and the labor force of the ‘‘stationary state’’ levels compatible with the Malthusian equilibrium level of per capita consumption, c . From (2) we can obtain the value a of the land-labor ratio that defines y as
166
Historical Applications of Heckscher-Ohlin Theory
y ¼ yða Þ
ð11Þ
We have therefore determined the long-run stationary values of the intensive magnitudes c , y , and a , but we yet have to determine the absolute levels of Y, C, A, and P. Land and the Frontier The marginal productivity of land and hence the rent per acre is determined in the stationary state as r ¼ y 0 ða Þ
ð12Þ
We assume that an acre of land deteriorates in fertility at a rate of m unless it is maintained. Suppose that the constant rate of time preference in the economy, and hence the rate of interest, is equal to d. Then by the usual asset pricing formula, the price of an acre of land would be p ¼ y 0 ða Þ=ðd þ mÞ
ð13Þ
Figure 7.2 depicts the determination of p by the intersection of the demand and supply curves for the stock of land in the stationary state. The downward-sloping curve pðaÞ in figure 7.2 shows the demand price for an acre of land as a function of the land-labor ratio. Since y 00 ðaÞ < 0 by the ‘‘diminishing returns’’ property of the production function (2), the value of y 0 ðaÞ capitalized by the reciprocal of ðd þ mÞ is the nega-
Figure 7.2 Stock equilibrium in the land market
Demographic Shocks and the Factor Proportions Model
167
tively shaped curve pðaÞ in figure 7.2. The intersection with the vertical supply curve of land per unit of labor a in the stationary state yields p as in (13). We now turn to the determination of the actual size of the land area A of the economy in the stationary state. Once this is determined, of course, the levels of L , Y , P , and C can all be obtained as well since we know the equilibrium intensive magnitudes a , y , and c . The area of arable land A ultimately available in an economic system depends upon climatic or geographical factors, on the one hand, and the technology of land clearance and maintenance, on the other. Population, and hence the labor force, is endogenously determined within a Malthusian framework by the same factors and hence is not an independent variable in its own right. In figure 7.3 below we consider the arable land available, A, as a function of the labor La , that is needed to ‘‘establish the frontier,’’ that is, to ensure that A acres are available to the economy. To express it differently, La is the cumulative amount of labor that has gone into the clearance of the A acres of land currently available in the economy. We postulate that La ¼ La ðAÞ;
La0 ðAÞ > 0;
La00 > 0
ð14Þ
where La0 ðAÞ is the marginal labor cost of clearing an additional acre, which increases with the amount of land already cleared, as indicated by the second derivative being positive as well. This formulation illustrates the Ricardian idea that extending the margin of cultivation is an
Figure 7.3 Determination of the land frontier
168
Historical Applications of Heckscher-Ohlin Theory
increasingly costly enterprise. The convex function (14) is plotted as the convex graph of La ðAÞ in figure 7.3. In the long-run stationary equilibrium it will be true that w La0 ðAÞ ¼ p
ð15Þ
that is, that the real marginal cost of clearing an additional acre, the left-hand side of (15), must be equal to the equilibrium price per acre p . Note that w and p in (15) have already been obtained by (4) and (13) alone. This enables us to solve (15) for the unique value A of A that satisfies this equation. This solution is depicted graphically in figure 7.3 where the slope of the convex function La ðAÞ at A is the value of La0 ðAÞ that satisfies (15), given w and p as determined already. It is instructive to substitute for p from (13) into (15) to obtain the relation w ¼ y 0 ðaÞ½1=La0 ðAÞ½1=ðd þ mÞ
ð16Þ
The right-hand side of (16) can be interpreted as the ‘‘indirect’’ marginal productivity of labor in clearing additional land, the marginal productivity y 0 ðaÞ of which is capitalized by the ‘‘gross’’ discount factor, the reciprocal of ðd þ mÞ. Since w is also equal to its direct marginal product by (4), we have the ‘‘efficiency condition’’ for labor allocation that it requires its direct and indirect marginal productivities to be equal. Having now determined the absolute magnitude A of arable land and knowing already the equilibrium values a , y , and c of the intensive magnitudes, we can easily obtain the absolute values L , Y , P , and C of the labor force, total output, population, and consumption. We have thus fully determined the equilibrium values of all the variables in the model. The relation between Y, A, and L is depicted in figure 7.4. Given A the concave function shows the relation between Y and L determined by the production function (1). The ray from the origin has a slope equal to y , so that the intersection with the concave function determines L and Y . Employment in Land Clearance We have established the cumulative labor effort needed to provide the economy with an arable land area of A . However, since land is assumed to depreciate at the rate m, for example through reforestation
Demographic Shocks and the Factor Proportions Model
169
Figure 7.4 Determination of total output
or soil erosion, it must continuously be replenished by the clearance of new land for the stationary state to be maintained. The amount of new land cleared will be an increasing function of the price of land and a decreasing function of the real wage, equal to p and w , respectively, in the stationary state. Letting N denote the amount of new land cleared (as opposed to the cumulative total, A) and Ln the amount of current labor required for this task, in the stationary state we must have NðLn Þ ¼ mA
ð17Þ
and Ln ¼ La ðA Þ La ½ð1 mÞA
ð18Þ
Assuming land clearance to be an explicit economic activity, ‘‘profits’’ equal to ð p N w Ln Þ must be maximized, which requires that P ðqN=qLn Þ ¼ w
ð19Þ
which is precisely the condition (15) obtained earlier to determine A by equating the price p to the marginal cost of land clearance w La0 ðAÞ. (At the margin it does not matter whether we use A or N.) Where does LN come from? The number of production workers L is equal to aP by assumption, so the LN workers required for the land clearance come out of the remaining pool of ð1 aÞP. Thus the labor force is the sum of the production workers and the members of the ‘‘land-clearing brigade,’’ and if we add the ‘‘unproductive’’ members of society we get the total population.
170
Historical Applications of Heckscher-Ohlin Theory
What happens outside of the steady state? Corresponding to any point ðA; LÞ in the input space the production function determines rðaÞ and wðaÞ by the marginal productivities of each input. Capitalizing rðaÞ by the reciprocal of ðd þ mÞ and ignoring expected capital gains and losses we can always obtain pðaÞ for every rðaÞ. Taking pðaÞ and wðaÞ as given the land-clearing sector maximizes profits in determining the amount of land cleared Nðp; wÞ as an increasing function of p and a decreasing function of w as indicated by N ¼ Nðp; wÞ qN=qp > 0;
qN=qw < 0
ð20Þ
Since the point ðA; LÞ is arbitrary it is clear that Nð p; wÞ may exceed, equal or fall short of mA, allowing A to rise or fall over time depending upon the direction of the inequality. Dynamic Stability From a dynamic perspective the model can be represented by the following two-dimensional system of differential equations L_ ¼ L_ ðL; AÞ
ð21Þ
A_ ¼ A_ ðL; AÞ
ð22Þ
where the dots indicate the time derivatives dL=dt and dA=dt respectively. The equilibrium state of this system is reached when the values of L and A are such that (21) and (22) are both equal to zero. These values are precisely the L and A that we have obtained above for the longrun stationary state. What is investigated here is whether the system will move towards the point ðL ; A Þ when LðtÞ and AðtÞ, the state variables of the system are not equal to these equilibrium values. The system will be dynamically stable, that is, LðtÞ and AðtÞ will approach ðL ; A Þ, when the Jacobian matrix of the partial derivatives of (22) and (23) qL_ =qL qL_ =qA J1 qA_ =qL qA_ =qA evaluated at the equilibrium point ðL ; A Þ satisfies qL_ =qL þ qA_ =qA < 0
ð23Þ
ðqL_ =qLÞðqA_ =qAÞ ðqA_ =qLÞðqL_ =qAÞ > 0
ð24Þ
Demographic Shocks and the Factor Proportions Model
171
The first of these is called the ‘‘trace,’’ and the second the ‘‘determinant’’ condition. Together they ensure that the characteristic roots of the matrix are negative, and have negative real parts, as required for dynamic stability. We may now evaluate each of the four partial derivatives to obtain their signs. The rate of change of population, and hence the labor force, is given by the difference between fertility and mortality rates: L_ ¼ ½ f ðcÞ mðcÞL
ð25Þ
from equations (5) and (6) above. Noting that per capita consumption c is proportional to per capita output y as given by (10), we obtain the sign of the partial derivative qL_ =qL ¼ ½ f 0 ðcÞ m 0 ðcÞaðqy=qLÞ < 0
ð26Þ
Since f 0 ðcÞ is positive and m 0 ðcÞ is negative, the expression within parenthesis is positive. But qy=qL ¼ y 0 ðaÞðA=L 2 Þ < 0
ð27Þ
since average productivity per worker falls when L is increased with A constant, ensuring (26). Partially differentiating (25) with respect to A we obtain qL_ =qA ¼ ½ f 0 ðcÞ m 0 ðcÞay 0 ðaÞ > 0
ð28Þ
From the previous section we have seen that A_ ¼ Nðp; wÞ mA
ð29Þ
Both p and w are functions of A and L, so we have qA_ =qA ¼ ðqN=qpÞðqp=qAÞ þ ðqN=qwÞðqw=qAÞ m < 0
ð30Þ
since qN=qp > 0;
qp=qA < 0;
qN=qw < 0;
qw=qA > 0
and qA_ =qL ¼ ðqN=qpÞðqp=qLÞ þ ðqN=qwÞðqw=qAÞ > 0 since qp=qL > 0;
qw=qL < 0
ð31Þ
172
Historical Applications of Heckscher-Ohlin Theory
Figure 7.5 The dynamic stability of the model
The trace condition (23) is clearly satisfied, since both diagonal elements of the matrix J have been shown to be negative. Figure 7.5 is the ‘‘phase diagram’’ of the dynamic system (21)–(22). It shows combinations of the state variables A and L that leave dL=dt ¼ 0 and dA=dt ¼ 0 along each of the respective functions, with the equilibrium point being ðA ; L Þ where these two functions intersect. It is easy to see that the determinant condition (24) is satisfied if the slope of the dl=dt ¼ 0 function is steeper than the slope of the dA=dt ¼ 0 function, that is, ðqL_ =qLÞ=ðqL_ =qAÞ > ðqA_ =qLÞ=ðqA_ =qAÞ The dL=dt ¼ 0 function is a ray through the origin with a slope equal to a , the land-labor ratio that equates the fertility and mortality rates to maintain the Malthusian equilibrium. For any given value of A points to the left of the dL=dt ¼ 0 ray have a greater than a and hence c greater than c , so population, and hence the labor force, must increase. By the same reasoning L must increase for a given A from any point to the right of the dL=dt ¼ 0 ray. Similarly, for any given L and amount of A above the dA=dt ¼ 0 line will result in A falling vertically to it or rising toward it from any value of A below that line. The dA=dt ¼ 0 locus has a flatter slope than the dL=dt ¼ 0, as required by the determinant condition for stability. The state variables will move according to the pattern of the arrows in each of the four regions of the input space.
Demographic Shocks and the Factor Proportions Model
173
Demographic Shock An immediate exercise is the impact of a demographic shock such as the bubonic plague. If we start from the equilibrium point ðA ; L Þ in figure 7.5, the population and hence the labor force as well collapses instantly from L to LO , while A is initially fixed. The real wage rises above w while the rent per acre falls below r . Hence, the land price also falls below p . In terms of figure 7.2 the land-labor ratio moves to the right, to aO , so that p falls from p along the demand curve to the point corresponding to pO ðaO Þ. The fall in p and rise in w make the amount of new land cleared fall below mA , and so the area of land cultivated contracts. The rise in per capita output and consumption stimulates the recovery of population and the labor force so the state variables move down and to the right along the path indicated in figure 7.5. Once the dA=dt ¼ 0 line is crossed from above the rising price of land and the falling real wage result in the new land cleared exceeding the depreciation mA, and so the area cultivated rises along with the labor force back to the equilibrium point ðL ; A Þ. Comparative Steady States Suppose that there is Hicks-neutral technological progress in the production function (1). At the original land-labor ratio this will raise productivity and hence consumption per capita, so as to increase population growth. To keep dL=dt ¼ 0 it is clear that the land-labor ratio a has to fall. This implies a rotation to the right of the dL=dt ¼ 0 ray as in figure 7.6. The dA=dt ¼ 0 line remains unchanged and so the new equilibrium point ðA ; L Þ involves an increase in both the labor force and the area of cultivation, with a lower land-labor ratio. Initially A remains fixed at A , since both p and w rise in the same proportion. But the increase in labor lowers the real wage rate and raises the rent and the price of land, and thus induces an expansion in the area cultivated as well. An improvement in the cost of land clearance, due either to better technology or to climatic conditions, will shift the La ðAÞ function in figure 7.3 downward for each value of A, and reduce the slope La0 ðAÞ as well. Thus, for each value of L on the dA=dt ¼ 0 line, the area cultivated, A, will have to rise until pðaÞ has fallen and wðaÞ has risen sufficiently to equate the price to the marginal cost of land clearance. As we
174
Historical Applications of Heckscher-Ohlin Theory
Figure 7.6 Technological progress
Figure 7.7 A reduction of the cost of land clearance
can see from figure 7.7 this will result in A and L both higher than the previous values A and L . From (13) it also follows that a reduction in the rate of time preference d or the depreciation rate m of land will raise the value of the marginal acre and hence require an increase in A to leave dA=dt still equal to zero for each value of L, while leaving the dL=dt ¼ 0 function unchanged. The result is therefore the same as in figure 7.7. As a final exercise a leftward shift in the fertility function f ðcÞ will raise c and hence y and c . The result is to rotate the dL=dt ¼ 0 locus
Demographic Shocks and the Factor Proportions Model
175
Figure 7.8 A fertility reduction
to the left, which results in a new equilibrium with A less than A and L less than L , as in figure 7.8. Performance: The Byzantine Empire In the next three sections we will briefly examine the economic performance of the three systems, mainly with respect to population, labor force, cultivated area, and output, but also looking at other factors such as urbanization and trade as indicators of these developments. The intention is to see whether we can relate these outcomes to the implications of the model and what we know of the exogenous shocks that occurred in each case. As we have noted, the Byzantine Empire before the plague struck in 542 was at its peak in terms of territory and population, put at about 26 million by Treadgold (1997:278) on the basis of estimates by McEvedy and Jones (1980). It is easy to imagine, with Cyril Mango (1994:68–69), what must have happened to the economy: All normal occupations were interrupted, prices of goods trebled and quadrupled, starvation set in, fields were deserted and the remaining farmers were burdened with additional taxes on the non-productive land of their deceased neighbours. There can be little doubt that the plagues of the sixth century, combined with an unprecedented sequence of natural disasters were a factor, perhaps the determining factor in the collapse of urban life. For it is a fact (though
176
Historical Applications of Heckscher-Ohlin Theory
some historians still refuse to recognize it) that all round the Mediterranean, the cities, as they had existed in Antiquity, contracted and then practically disappeared.
By 610, as a result of the plague, the Byzantine population had fallen to 17 million, and it continued to fall to a mere 7 million by 780 after the loss of Egypt, Syria, Palestine, and North Africa to the Arabs and further outbreaks of the plague. The population of Constantinople, which was at least 300,000 at the time of Justinian, fell to a mere 40,000 by the middle of the eighth century, after another disastrous outbreak of the plague in 747, ‘‘probably . . . the lowest point in the medieval history of Constantinople’’ (Mango 1994:78). Antioch and Alexandria, the next largest cities after Constantinople, were lost to the Arabs in the seventh century. Thessalonica was the only other remaining city of any size. Several of the cities of Asia Minor declined to mere kastra, citadels surrounded by small civilian settlements (Mango 1994; Cameron 1993). That this development should not have been related to the appearance of the plague is difficult to believe. Even an author as cautious as Averil Cameron (1993:164) is quite emphatic: The effects are certainly hard to quantify . . . but it is hard not to think that plague must have been a factor in undermining the generally thriving state of cities in the Near East in the early part of the sixth century . . . . Historians vary sharply in the amount of weight that they are willing to attach to the sixth- and seventh-century plague. Yet the fact remains that this seems to have been the first appearance of bubonic plague in Europe; its impact must therefore have been far greater than that of the regular diseases which ravaged ancient cities as a matter of course.
The period 641–780 is described by Treadgold (2002) as the ‘‘Struggle for Survival’’ against the twin enemies of the Arabs and the plague. Byzantium may have escaped total collapse by a slim margin. State revenue fell drastically as people died (Treadgold 2001:67): The evident explanation for this decline in state spending is the same as that for most of Justinian’s failures: the effects of the plague after 541. These were so different from most historical events in magnitude and kind that they were inevitably underestimated by most Byzantine observers, as indeed they have been by most modern historians. Not even the return of the plague in the thirteenth century wreaked such havoc, because it recurred less often and reduced a denser population. If the first outbreak had killed Justinian, as it almost did, it might well have brought on the fiscal and military collapse of the empire that he barely adverted.
It was bad enough as it was.
Demographic Shocks and the Factor Proportions Model
177
While revenue and population shrank, the size of the army did not fall in the same proportion, so a greater share of resources was devoted to the military, and a reorganization of society along military lines, in so-called themata (military districts, each with its own army), took place. As our model predicts, however, per capita income appears to have been on the rise and income disparities were being reduced. Treadgold (2002:150) states that in ‘‘comparison with earlier Byzantine times the rich appear to have been poorer and the poor richer,’’ which is consistent with the rise in the land-labor ratio lowering rents and raising wages, in terms of our model. He also observes that peasants benefited from the greater availability of land. Labor had become a scarce factor and the peasants benefited from this. It became difficult to tie them to the soil. Their freedom of movement increased and they had ‘‘complete and unlimited legal disposal over . . . [their] land. The Byzantine sources show quite clearly that peasant land was handed down from generation to generation by inheritance and that it could be freely alienated by the possessor just as he chose—by sale, by gift, or limited lease’’ (Ostrogorsky 1966:210). The Byzantine Empire had uncultivated land at its disposal, in its very heartland, Asia Minor, where the military holdings granted out in return for military service could not obtain labor without bringing in foreigners, mostly of Slavonic origin. The rise in per capita income that all this seems to indicate could also explain how a greater than proportionate military burden was borne, relative to the pre-plague years. In relation to our model the situation would be ripe for the rise in per capita consumption to induce a recovery of population and an associated increase in the area under cultivation. While the empire had survived, its prospects around the year 800 did not look too good in the ongoing tripartite struggle for power against its two great rivals, the Carolingian Empire of the West and the Abbasid Empire of Haroun al-Rashid in Baghdad. Byzantium had to contend not only with the Arabs but with formidable barbarian peoples on its frontiers, such as the Bulgars and the Pechenegs. In the west the aggressive Normans intruded successfully into Byzantine possessions in southern Italy and the Balkans, after taking Sicily from the Arabs (Norwich 1991, 1997). Population, and with it manpower and revenue, began to revive from 7 million in 780 to 12 million in 1025 on the death of the Macedonian Emperor Basil II, who crushed the Bulgars and stabilized the Balkan frontiers. Also significant, particularly for the future, was the
178
Historical Applications of Heckscher-Ohlin Theory
conversion to Christianity of the Kievan Rus ruler Vladimir and his marriage to a Byzantine princess to cement a mutually beneficial alliance. Meanwhile the Carolingian Empire was fragmented in the succession struggles between Charlemagne’s heirs and the power of the Sunni Abbasid caliphs was challenged by the Shia Fatimid caliphate in Cairo, while Byzantium preserved its political and religious unity. As the population recovered, the man-land ratio increased, and the relative bargaining power shifted in favor of the larger landholders, who proceeded to buy up peasant property. Efforts on part of the rulers to protect the peasants proved to be of little avail (Ostrogorsky 1966:221–222): After the death of Basil II [in 1025] the long series of these laws came to an end. . . . For, as even the government regulations of the tenth century, in spite of their extreme severity, had been unable to suppress the buying-up of peasant and military lands, now the passively benevolent attitude of the government meant that the great landowners’ capacity for expansion could develop to the full. The destruction of the small freehold properties continued unrestricted; the great landowners absorbed the land of the peasants and soldiers and made the owners their serfs. . . . Certainly there were free peasants in the late Byzantine period; but whereas in the middle Byzantine period, from the seventh to the beginning of the eleventh century, the free and moving peasantry is the chief factor in agrarian development and the backbone of Byzantine agriculture, from the eleventh century onward, just as in the early period, the great landlord dominates the scene. The agrarian history of the late Byzantine period is that of great landowners and their dependants.
Just when it seemed that the empire could continue its revival a fateful new challenge arose with the emergence of the Seljuk Turks. They took over much of the interior of the Anatolian Plateau after defeating the Byzantines at Manzikert in 1071. This shifted the center of gravity of the empire from the interior of Asia Minor to the coastal regions of the Balkans and the Black Sea. The rising commercial and naval power of Venice, Pisa, Genoa, and the other Italian cities was accommodated by the Byzantines with increasing diplomatic concessions, such as exemptions from customs duties and ‘‘extraterritorial’’ rights in Constantinople (Nicol 1988). The relative expansion of the West was strikingly displayed in the Crusades, which liberated Jerusalem and Antioch and established new kingdoms in the Holy Land (Runciman 1951; RileySmith 1987). These Christian allies did not accept the authority of Byzantium, and the Fourth Crusade of 1204 that sacked Constantinople and established the so-called Latin Empire that lasted until 1261
Demographic Shocks and the Factor Proportions Model
179
was a fatal wound from which the empire never recovered until its fall at the hands of the Ottoman Turks in 1453 (Queller and Madden 1997). Despite the economic expansion from the tenth century onwards, part of which has been well documented by Harvey (1989), it is difficult to resist the impression that the economic history of the Byzantine Empire was one of protracted demographic collapse, due to plague and the Arab invasions, from about 550 to 780, followed by a long recovery during which power systematically shifted in favor of the large landowners, set back by the intrusions of the Fourth Crusade and the Turks. Something akin to a new demographic-economic equilibrium was being restored, within the restricted post-plague area imposed by the Muslim advances, as would be implied by our model. This restoration is symbolized by the fact that Constantinople seems to have had the same population, about 300,000, at the time of Justinian and the onset of the plague, as it had when it was sacked by the Crusaders in 1204, after having fallen to as low as 40,000 in the middle of the eighth century. This was also about as many inhabitants as it had on the fateful day of May 29, 1453. There does not appear to have been any notable technological progress or other economic innovation in the entire history of the empire: In general, Byzantine technology was extremely conservative. Byzantine agricultural implements remained virtually unchanged from Roman times. The peasant continued to use the light plow dragged by a pair of oxen. It was made of wood and had a removable iron plowshare; it did not have wheels, so the plow bit rather than cut the soil. The scythe was not in use in Byzantium, and the image of Death with its scythe in hand, so popular in the West, would have left the Byzantines unmoved. (Kazhdan and Epstein 1985:27)
The large estates remained partly uncultivated, very much due it seems to faulty techniques (Ostrogorsky 1966:211): The difficulty of making proper use of the larger estates was partly due to the primitive conditions of economic technique; for in this respect the Byzantine Empire, so far ahead in culture, was in many ways far behind the West. Thus Byzantium to the end of its days continued to employ an extremely uneconomic and antiquated harness for draught animals, while by the tenth century the West had evolved a greatly improved method of harnessing.
Byzantium maintained and adopted the administrative, military, and economic institutions that it had inherited from antiquity, but did not make any significant new breakthrough, as far as one can gather, other than ‘‘Greek fire,’’ the lighted naphtha that was used successfully
180
Historical Applications of Heckscher-Ohlin Theory
in naval battles against the Arabs in an early form of chemical warfare, and the importation of the silkworm from China in the sixth century in an early example of commercial espionage. Characteristically, this led to a significant but ‘‘luxury’’ industry, which was also imitated and supplanted in due course by both of its great rivals, the Muslims and the Latins. The critical frontier, in the case of Byzantium, was always a political and military one, not one of settlement and cultivation. As Harvey (1989) points out, there was considerable land reclamation accompanying the population growth from 900 to 1200, but this is not likely to have constituted any net addition to the cultivated area in the same regions before the onset of the plague. None of this should be taken as denigrating the great resilience and adaptability that the empire displayed in dealing with an almost uninterrupted succession of military and political challenges and the crucial role in world history that it played in relation to the Russians and the other Slavic peoples of Eastern Europe. The Islamic World The extent and rapidity of the Arab conquests of the seventh and eighth centuries is one of the most remarkable features of world history. ‘‘The speed, magnitude, extent and permanence of these conquests excite our wonder and almost affront our reason, but the historian who seeks to explain them is impeded by the deficiency of the evidence at his disposal,’’ writes J. J. Saunders (1965:39). The first conquests, however, can perhaps be partly explained with the aid of the Findlay (1996) model of the extension of empires, or simply by interpreting the extension of the land frontier in the present model as a military operation. Justinian’s attempt to reconquer the territories lost by Rome to the barbarians had turned the Byzantine frontier into a triple war frontier: Italy, Africa, and the east. The eastern frontier was weakened by the necessity to maintain troops in the two other war theaters. The eastern war led to nothing, and the plague outbursts in the sixth and seventh centuries in the eastern Mediterranean and the western part of the Sassanid Empire left the two contending empires militarily weakened, with little power of resistance, whereas the Arabs had escaped the deadly disease and could take advantage of the population vacuum. Damascus, Jerusalem, and Alexandria fell to Islam in
Demographic Shocks and the Factor Proportions Model
181
rapid succession between 635 and 642, and farther east the Arabs simultaneously penetrated Sassanid territory. Islam, and Arabic, the language in which the Qur’a¯n was written, imposed a cultural unity over a vast extent of diverse peoples and physical environments, stretching from the shores of the Atlantic to the oases of Central Asia and the mountains of Afghanistan. This vast territory was never a single politically and administratively unified empire but it did maintain a unique identity as the Dar al-Islam, or Abode of Islam, as opposed to the Dar al-Harb, or Abode of War, that is, the infidel. Thus most of the welter of dynastic changes and power shifts that form so much a part of medieval Islamic history can be ignored for purposes of this paper since they mostly took place within the same system. It was only on the frontiers of Spain and in the Mediterranean islands, before the Crusades, that land and people moved in or out of the Islamic world itself as a result of political conflict. In Central Asia much of the conversion of the nomadic Turkish and later Mongol tribes and states took place peacefully. The Crusader kingdoms in Syria and Palestine were exceptions, but only temporarily, since they were eventually returned to the Islamic fold by Saladin and the Mamluks. In the case of Islam, the extension of the frontier of cultivation was a military operation rather than a peaceful land clearance affair (Hitti 1970:144): Islam did provide a new battle-cry, a convenient rallying-point and a party watchword. It undoubtedly acted as a cohesive and cementing agency for the heterogeneous masses never before united and furnished a large part of the driving force. But it is hardly in itself enough to explain the conquests. Not fanaticism but economic necessity drove the Bedouin hordes, and most of the armies of conquest were recruited from the Bedouins, beyond the confines pf their arid abode to the fair lands of the north. The passion to go to heaven in the next life may have been operative with some, but the desire for the comforts and luxuries of the civilized regions of the Fertile Crescent was just as strong in the case of many.
Fertile soil was available in abundance in the conquered territories, in the valleys of the Guadalquivir, Nile, Euphrates, Tigris, Oxus, and Jaxartes rivers, with the Indo-Gangetic plain being added by the Delhi Sultanate early in the thirteenth century. The string of oases from North Africa to Central Asia also yielded high returns. On the whole, however, the Abode of Islam was situated in one of the most arid
182
Historical Applications of Heckscher-Ohlin Theory
zones of the civilized world, and ‘‘the struggle between the desert and the sown’’ has been a persistent theme of its existence. To quote the eminent French geographer Xavier de Planhol (1959:102), ‘‘the heart of Islam remains that desert zone which, slanting across the globe from the Atlantic Ocean to Central Asia, includes the whole of the ancient world between, on the one side, the humid zones of intertropical Africa and monsoon-moistened Asia and on the other side the wet and temperate climate of Europe.’’ In this zone, according to de Planhol (1959:124), the towns, linked to each other along the trade routes, dominate their rural agrarian environs, which, despite their density, are often no more than ‘‘a farming or truck-gardening suburb.’’ Reflecting its origins in Mecca and Medina, Islam in this view is essentially a religion of traders and city-dwellers. It has always had difficulty penetrating mountainous and forested areas. Pastoral nomadism is well adapted to the environmental conditions of the Islamic zone, and Bedouin Arab, Berber, and Turkish nomadic tribes have all interacted positively and negatively with the sedentary cultivators and townsfolk over the centuries, providing the fourteenthcentury Tunisian sage Ibn Khaldun (1958) with the theme of his great work. Interestingly, Khaldun says (1958, 1:302), ‘‘Arabs can only conquer flat territories’’ and also that (1958, 1:308) ‘‘desert tribes and groups are always dominated by the urban population.’’ Jared Diamond (1997) has popularized the idea that agricultural innovations can be diffused more readily across the same latitudes, since climatic conditions are similar, rather than on a north-south axis. With its largely ‘‘horizontal’’ extent around the globe, within a relatively narrow band, the early Islamic world provides an excellent example of this thesis. As Andrew Watson (1983) has demonstrated convincingly in a seminal work, the first three or four centuries of Islam were marked by a remarkable agricultural revolution that saw a very wide variety of new crops diffused from its eastern margins in India all the way to Morocco and Spain. These included such major crops as cotton and sugarcane, as well as rice, hard wheat, sorghum, citrus fruits, coconuts, bananas, artichoke, spinach, and eggplant. The introduction of these crops required complementary efforts in irrigation, the construction of canals and the opening up of new lands. The agents of such change were not only farmers and landowners but also rulers and officials who provided the necessary institutional infrastructure. The following quotation (Watson 1983:129) describes a process that corresponds exactly to the implications of our model:
Demographic Shocks and the Factor Proportions Model
183
The agricultural revolution was bound up with an ill-documented but none the less real demographic revolution which seems to have touched most parts of the Islamic world from roughly the beginning of the eighth to the tenth century. Rising population levels and increasing levels of output of foodstuffs must continuously have interacted: though both were affected by other factors, at times demographic growth must have been the result of agricultural progress, at times its efficient cause.
Watson cites many instances of this agricultural expansion from all over the Islamic world, and the reader is referred to his book and the many sources cited therein for the evidence. Unfortunately, there appear to be no numbers. Perhaps the strongest evidence for the agricultural revolution of the early Islamic world is the extent and scale of its urbanization. Eric Wolf (1966:13) has defined the peasantry ‘‘in terms of its subordinate relationships to a group of controlling outsiders,’’ and, conversely, urban areas are dependent on the surrounding countryside for their growth. Watson (1983:132) challenges many published estimates of city size as biased downward, by sheer ‘‘Orientalist’’ prejudice or ‘‘on the grounds that it is more responsible to underestimate than to overestimate.’’ Baghdad and Samarra, the two capitals of the Abbasid dynasty, could have had populations of close to a million, although half a million would be more plausible and impressive enough, since Constantinople at the time would not have exceeded 300,000. Basra seems to have had at least 200,000. One colorful citation (Watson 1983:131) says that ‘‘along the Tigris settlement was continuous, so that before dawn crowing cocks answered one another from housetop to housetop all the way from Basra to Baghdad.’’ On the eve of the Black Death, the twin city of Fustat-Cairo also had a population that was of the order of half a million, and Nishapur in the ninth century is put at between 100,000 and five times that number. In the west Fez and Qairawan had several hundred thousand each but Cordoba at its peak was comparable to Baghdad at its height, estimated to be between half and one million. Cordoba also surpassed the one million mark. According to a census taken towards the end of the tenth century, the city had 1,600 mosques, 213,077 houses occupied by the lower and middle classes, another 60,300 inhabited by the higher bureaucracy and the aristocracy, and 80,455 stores (Ocan˜a Jime´nez 1975:47). The Islamic world also had very many cities of lesser rank, far outstripping Byzantium and Western Europe.
184
Historical Applications of Heckscher-Ohlin Theory
Agriculture clearly had to be very productive to support urban settlements on such a scale, (as well as the Arab cultural advance in general). The relationship was not confined simply to food supply. Major processing industries such as sugar refining and textile manufactures relied on the supply of raw materials from the agricultural sector. In addition to cotton, raw silk and hemp for linen were important inputs to urban textile industries. Many of these manufactures were exported to other parts of the Islamic world, and also to Europe. In addition to these high value-added manufactured exports, the Islamic world, through the Red Sea and to Persian Gulf, was also the entrepoˆt for the precious spices from India and Southeast Asia to be distributed to Europe. Looking at the admittedly unsatisfactory data provided by McEvedy and Jones, we see that the population of the Islamic world seems to have grown from about 21 million to about 27 million from 700 to 1000, say about 30 percent over three centuries. Issawi (1981) gives 28– 36 million for 750 and 35–40 million for 1000. Taking the 28 million low estimate for 750 and the 40 million high estimate for 1000 we get what would be a maximum increase of 43 percent for two and a half centuries, as compared with 30 percent for three. While noticeably better than the performance of Byzantium at about 25 percent over three centuries, even the 43 percent over 250 years can hardly be called a ‘‘demographic revolution.’’ Thus, while fully agreeing with Watson’s picture of complementary agricultural and demographic expansions in the early Islamic world, we should not exaggerate the magnitude of either. There is little doubt that what held back the further progress of early Islamic agriculture was the unfavorable nature of the physical environment and climatic conditions under which it operated. Once the power of conquest was gone, the character of the frontier changed from one of war to one of land clearance. In the ‘‘struggle between the desert and the sown’’ the sown had three good centuries, from 700 to 1000, but then the desert seems to have begun its counterattack. Islam was thus hemmed in by its unfortunate geographical circumstances. As Watson says, there is the possibility that the earlier expansion may have become over-extended, with excessive reliance on irrigation leading to soil erosion. Settlement was not continuous in many areas, raising costs of communication. Warfare and the breakdown of central authority, and the raids of the ever-present nomads, also shrank the margin of cultivation. The social
Demographic Shocks and the Factor Proportions Model
185
institutions of the waqf, donation of land for religious or charitable purposes, and the iqta, a revocable military fief, also were not conducive to sustaining agricultural productivity. The McEvedy-Jones population estimates decline from about 27 million in 1000 to about 23 million in 1300, reflecting the loss of cultivable area and the cessation of innovation. The impact of the Mongol invasion (Findlay and Lundahl, in press), the sacking of Baghdad in 1258 by Hu¨legu¨ and the annihilation of the Abbasid Caliphate, was also of course an additional major factor in this regard. Western Europe Western Europe at the outset of our period had a Latin Christian core, soon to be flanked by a Muslim Spain in the west, with a pagan fringe in the Scandinavian north and in the east, with the Frisians, Saxons, Balts, and Slavs inhabiting the shores of the North Sea and the Baltic and the fertile plains beyond the Elbe, the Oder, and the Vistula. By the end of the period the pagan fringe had long since been almost entirely converted and the Muslims in the Iberian Peninsula had been pushed back all the way south to the narrow confines of the kingdom of Granada, and driven out of Sicily, Crete, and other Mediterranean islands that they had temporarily occupied. Which were the factors behind this extraordinary advance, which the Roman Empire at its height was not even able to attempt, much less achieve? Lynn White (1962, 1978) had a deceptively simple answer, that could be colloquially phrased as ‘‘it was the stirrup and the plow, stupid,’’ which led to his being roundly condemned by Sawyer and Hilton (1963) for the intellectual crime of ‘‘technical determinism.’’ The plow in question was the heavy wheeled plow with an iron coulter and plowshare, drawn by a team of horses or oxen, that cut deep into heavy soil in long straight furrows, turning the soil over with a moldboard. Though apparently known in antiquity it was not employed, being unsuitable for Mediterranean conditions, where the much lighter so-called scratch plow prevailed. The plains of northern Europe, however, were ideally suited to the heavier instrument. Horses also proved to be more efficient draft animals than oxen, after the invention of the horse collar multiplied the load that they could pull by up to a factor of five. The wear of the heavy loads on the horse’s hooves required the complementary invention of the nailed horseshoe.
186
Historical Applications of Heckscher-Ohlin Theory
Another innovation, the three-field rotation in place of the more traditional two-field rotation, raised productivity per worker by 50 percent, if additional land were available, clearly providing a stimulus to land reclamation on a vast scale. In the two-field system, half the acreage was sown with winter crops, while the other half was left fallow. This made sense in the Mediterranean area where rainfall was concentrated to the winter, but in northern Europe, where sowing in the spring could bring another crop, one-third of the area was sown in the fall (wheat or rye), another third in the spring (barley, oats, or legumes), and the last third was left fallow. The following year the second field was sown in the fall, and so forth, until the cycle recommenced. The three-field system both increased the area under cultivation and the range of crops grown. The new system was introduced in France during the eighth century, but took time to spread, and it was not until after 1250 that the speed of diffusion became rapid. Even so, the two systems coexisted during the thirteenth and fourteenth centuries (Grigg 1980:73). This new technology required changes in the social organization of the peasant community to adopt a more cooperative basis, in order to take advantage of the economies of scale offered by the more expensive plow and draft animals. The plow is a labor-saving device. It was introduced by a society that had undergone a population decline, but below a certain ‘‘threshold farm size’’ the introduction of the plow is not profitable, and when the cost of feeding the draft animals is taken into account this size increases (Lundahl 1979:590–591).4 The need for oats to feed the horses stimulated greater variety in the mix of crops. These complementarities led to a highly productive system of mixed farming with cereal cultivation and livestock raising becoming the predominant form of agriculture in much of Western Europe. The stirrup, unknown in antiquity despite its apparent simplicity, led to a revolution in warfare with the armored knight on his powerful warhorse relying on the lance to pulverize opposition. The high cost of horse and armor, and the long period of training required for knightly proficiency, would have been impossible without the surplus for rents and taxes over the peasant’s consumption that the heavy plow provided. At a time such as the ninth and tenth centuries, with incessant raids by the Muslims in the south and west, Vikings on the northern coasts, and marauding Magyars from the east, the peasant also needed the protection of the knight, so that ‘‘stirrup and plow’’ together were
Demographic Shocks and the Factor Proportions Model
187
the effective cultural package powering the rise of Western Europe during these centuries (Bloch 1962). Both peasant and warrior needed the consolation and spiritual guidance of the priest, who in turn was supported and protected in his cloister by the peasant and the warrior, together constituting the ‘‘Three Orders’’ of medieval feudal society as depicted by Georges Duby (1980). Though the concept of the frontier is usually associated with Frederick Jackson Turner (1986) and the westward expansion of the United States in the nineteenth century, it is if anything even more applicable to early medieval Europe, as many authors have realized and pointed out in the excellent study edited by Bartlett and Mackay (1989). There clearly was an internal economic frontier in already settled areas, between stretches of arable land and the immense forests surrounding them. When reflecting over the data provided by the Domesday Book,5 M. M. Postan (1966:549) unequivocally concludes that the ‘‘Domesday facts clearly denote dense, and hence ancient settlement, the product of at least six centuries of internal colonization.’’ The colonization pattern was generalized in Western Europe (North and Thomas 1973). Colonization took on different shapes. Trees and bushes were cleared in the neighborhood of villages or in the waste further away, contributing to the scattering of fields. New settlements were established in the forests between villages. Coastal areas were reclaimed by the construction of embankments, and marshes were drained. Most important was the advance into forests. During the first centuries after the plague, ‘‘the forest seems to have held sway over the whole natural landscape’’ of Western Europe, as Duby (1974:5) writes. Lowland forest was being cleared in the eleventh and twelfth centuries both in England and France, and at the end of the twelfth century the process continued into upland areas as well, in Brittany, above the Moselle, in the Vosges, the Alps, and the Pyrene´es (Grigg 1980:71–72). There also was an external military frontier, which required armed force to push back Muslim warriors in Spain during the entire Reconquista, or Saxon and Slav pagan tribes east of the Elbe, particularly after 1150, settling and cultivating the lands thus won and either converting, expelling, or killing the previous occupants. With land abundant and labor scarce, particularly on the eastern frontier, once the Saxons or Slavs accepted German rule, Jesus Christ as the ‘‘Deus Teutonicus,’’ and the German heavy plow in place of their less efficient wooden one, they were valued and protected subjects of the German
188
Historical Applications of Heckscher-Ohlin Theory
princes and bishops. After a few generations the population became culturally ‘‘German’’ despite the abundant and perhaps even dominant presence of Slavic genes. On the other hand, Polish and other Christian Slavic rulers also welcomed the efficient German peasants as settlers on their own lands. Assimilation was more difficult in Spain, where cristianos nuevos were regarded with considerable suspicion, and even in Ireland where there was tension between the native Celts and the Anglo-Norman feudal lords who were granted estates soon after the conquest. The complementarity between the stirrup and the plow and the three orders on both the eastern and western frontiers is revealed in the vital role played by the military orders in both cases, the Teutonic Knights in the east and the Orders of Calatrava and Santiago in Spain. The Reconquista in Spain, with Santiago Matamoros and El Cid as its foremost symbols, and the wars against the Balts and Slavs, merged into the crusading mentality that swept Christendom, and also saw the excursions against the Albigensian heretics in Provence and into Syria and Palestine against the Muslims under the banner of the cross. Not surprisingly, the movement had permanent effects within the contiguous frontiers on the mainland of Europe but proved ephemeral in the Holy Land. The role of Christianity in the extension of the frontier was not confined by any means to the inspiration of crusaders. As the greatest landowner in Europe the church took an active part in promoting land clearance and improving agricultural productivity in other ways. It has also been pointed out, by both Lynn White and Georges Duby, that Christianity induced a more instrumental and exploitative view of nature as something purely for the service of man, rather than as being imbued with a spirit of its own that had to be respected and propitiated. The great forests of beech and oak that were cut down to clear land for pasture and tillage were the abode of spirits that the animistic pagans worshipped. This cultural shift was thus no less important than the heavy plow or the three-field rotation in making the agricultural revolution possible, so the charge of ‘‘technical determinism’’ by White’s critics is not entirely fair. Indeed most of their objections are on matters of the speed and scope of the technical changes and not on their ultimate significance. Hilton (Sawyer and Hilton 1963) says that rather than the causal sequence from plow and three-field rotation to food supply to population it might be the other way around, from exogenous increase in population to the need for more food and hence to
Demographic Shocks and the Factor Proportions Model
189
the changes in agricultural technique, anticipating a thesis advanced by Ester Boserup (1965). One advantage of the modeling approach adopted in this paper is that it enables us to escape from being trapped in chicken-egg arguments of this type. In our model both food supply and population are endogenous variables, depending upon the state of technical knowledge, geographic and climatic conditions, and the biological and behavioral determinants of fertility and mortality rates. The heavy plow may have been known to antiquity, and for that matter to the Islamic world, but there was no incentive to adopt it under Mediterranean environmental and social conditions, as White himself clearly says. The results in terms of population growth in any case were spectacular, as seen from the figures already cited. The McEvedy-Jones numbers further indicate that growth was particularly rapid in the areas of modern Germany and Poland, where it trebled over the period from 1000 to 1340 and also in the areas of modern France and Great Britain. Italy doubled from 5 to 10 million over the same period, but even this was below the average for Europe as a whole. These patterns of population growth in the old and new areas of Europe are consistent with the frontier thesis adopted in this paper. Hand in hand with the increase in area cultivated and population went an increase in the growth of towns and trade. While in 1000 there were only a hundred places in Europe that could be called towns, and half of them were in Italy, three hundred years later the figure had increased to between 4,000 and 5,000 (Grigg 1980:77). Western European cities, however, were far smaller than their Islamic counterparts. In 1292, Paris had a population of no more than 59,000 people, and as of 1328 the figure had possibly increased to 80,000. Toulouse, an example of a large provincial town, around the same time presumably had no more than 24,000 inhabitants. In Italy, where the larges cities were found, Milan had 52,000 in the thirteenth century, Padua 41,000 in 1320, Naples 27,000 in 1278, Venice 78,000 in 1363, Bologna 32,000 in 1371, and Florence 55,000 in 1381. Before the middle of the fourteenth century only Milan, Venice, Naples, and Florence, and possibly also Palermo, exceeded 50,000 inhabitants. Some of these cities may have approached the 100,000 mark. In the Low Countries, Ghent had some 56,000–60,000 inhabitants in 1356, and Bruges 25,000–35,000 in 1340. Antwerp was a small town of 18,000 in 1374. In England, London dominated completely, with 35,000–45,000 in 1377 followed by the far smaller York and Bristol, with figures somewhere between 10,000 and
190
Historical Applications of Heckscher-Ohlin Theory
14,000. Barcelona, Cordoba, and Seville (plus Granada in the Moorish zone) were the only cities in Spain with more than 40,000 inhabitants. Only Cologne in Germany was close to the 40,000 mark, while Metz, Strasbourg, Nuremberg, Augsburg, Vienna, Prague, Lu¨beck, and Magdeburg had populations around 20,000 (Russell 1958, tables 63–65; van Werveke 1963:38–39). Even with a generous allowance for the influence of the Black Death on the later figures the European cities were thus extremely small, relative to Constantinople and the Islamic world. Trade took place not only within Europe but with the Byzantine and Islamic worlds as well, leading to the well-known identification by Robert Lopez (1976) of a ‘‘commercial revolution’’ of the thirteenth century that was also augmented by the growing trade with Asia both by sea and overland due to the Pax Mongolica. As is well known, a Genoese ship from the Black Sea port of Kaffa engaged in this lucrative trade returned to the Mediterranean in 1347 and opened another act in the drama of ‘‘Rats, Lice and History’’ (Zinsser 1935). Even before the onset of the Black Death, however, the preceding century was marked by what Archibald Lewis (1958) termed ‘‘the Closing of the Medieval Frontier 1250–1350.’’ The internal frontier was reaching its natural limits within the prevailing technology and there is the possibility that climate may have started to become less favorable. Tillage competed with pasture to the detriment of the proper balance between the two. Population continued to grow while the supply of land was not keeping pace, lowering wages and peasant incomes while raising rents (North and Thomas 1973:48), leading to growing inequality and social conflicts in both the countryside and the towns. The external frontiers of Europe also ceased to expand, with the Muslims recovering the Holy Land and the Byzantines Constantinople in 1261. The second decade of the fourteenth century saw the outbreak of what William Jordan (1996) has called the ‘‘Great Famine’’ of 1315–22. A population of Europe in excess of 70 million, under the prevailing technology, was getting to be unsustainable within its geographical confines. A Malthusian crisis of major proportions was clearly looming. Population densities in certain rural areas in 1300 had increased to the point where they were comparable with those of the early nineteenth century, when the farming technology was vastly superior (Grigg 1980). David Grigg (1980:82) summarizes: For perhaps two centuries the expansion of the cultivated area and the adoption of new techniques was sufficient to keep production up with population
Demographic Shocks and the Factor Proportions Model
191
growth. But by the middle of the thirteenth century the supply of agricultural land was running out; it was in this century that there are most signs of attempts to intensify production by growing legumes and reducing the fallow but they seem to have been insufficient. The primary blockage to improving yields was the lack of livestock manure, and in the densely populated arable areas of south-eastern England and northern France this reflected the lack of grazing, as population growth led to the ploughing of grazing land.
Western European agriculture was getting into a situation typical of a number of today’s developing countries, where the pressure of population leads to increasingly intensive cultivation, with diminishing returns to labor and falling productivity per acre, tantamount to a reduction of the effective land area, in a sequence that easily feeds itself (Lundahl 1979). The Genoese ship from Kaffa that brought the plague into the West was in the nature of a historically necessary accident waiting to happen. Conclusions In the middle of the sixth century AD, the Mediterranean world was struck by the so-called Plague of Justinian, an epidemic that would recur in further successive waves until the mid-eighth century. No good estimates exist with respect to death tolls, but there is not the slightest doubt that in the areas where it hit, the consequences were extremely severe. The diffusion in space was limited, however. Northern and northwestern Europe apparently escaped, and so did the areas that a century later would constitute the core areas of the new world religion: Islam. As a result of the plague, the population of the Eastern Roman Empire hit a trough some time around the late eighth century and then began to rise until the advent of the Black Death in the midfourteenth century. Simultaneously, the Islamic population rose until the Mongol invasions in the mid-thirteenth century. Europe, in turn, saw a population decline in the wake of the plague, followed by a rise that accelerated after 1000 until the early fourteenth century, when famine was followed by another outbreak of the plague. Thus, all three geographical territories display a common demographic pattern: a decline in the size of the population, followed by a rise, continuing on beyond the previous peak. As our essay demonstrates, this pattern can be made subject to analytical representation in the form of a simple Malthusian model where
192
Historical Applications of Heckscher-Ohlin Theory
birth and death rates are functions of per capita consumption, production is a function of labor and land, and the extent of the arable area determined by the existence of an agrarian or military frontier that can be extended at a rising cost in terms of labor. This simple model can be used to investigate the impact of a demographic shock. The shock leads to an instantaneous rise in the landlabor ratio due to the collapse of population, higher wages and lower rents, and a gradual reduction of the area under cultivation. Per capita consumption, among the survivors, increases and stimulates population growth. Due to the Malthusian characteristics of the model, however, this increase cannot continue, so the land-labor ratio must fall. The price of land increases and induces land clearance at the frontier and an expansion of the area under cultivation until the original levels of both land and labor are restored. The model can also be used to investigate the effects of a discrete technological change. This raises labor productivity and consumption per capita and so makes the population grow. Population growth, however, cannot go on indefinitely, and in order to make it cease the landlabor ratio must fall. The larger population raises rents and the price of land, inducing an increase of the area under cultivation but less than proportionately to the increase in population, thus bringing about the fall in the land-labor ratio required to restore the Malthusian equilibrium. The extent of the resulting increases in population and the supply of land depend upon the magnitude of the technical change and the elasticity of the endogenous land frontier with respect to the rise in the price of land that it induces. A reduction of the cost of land clearance, as a result of technological progress or more favorable climatic conditions, reduces the amount of labor necessary to sustain a given land area as well as the labor needed to clear an acre of new land. The area under cultivation has to increase until the land price has fallen enough and the wage rate risen enough to equate the land price and the marginal cost of land clearance. This, in turn, also increases the population and the labor force, in the same proportion as the increase in the supply of land, to preserve the Malthusian equilibrium. Analogously, decreases in the rate of time preference or the exogenous rate of land destruction increase the price of land at the margin and hence also the area under cultivation and the labor force in the same proportion.
Demographic Shocks and the Factor Proportions Model
193
Finally, an autonomous fertility reduction leads to higher per capita consumption and production in the Malthusian equilibrium. The labor force shrinks, and with it also the area under cultivation, but less than in the same proportion, permitting the economy to enjoy a permanently higher standard of living as a result of the greater relative landabundance induced by the exogenous reduction in fertility. The model thus makes some explicit predictions, notably with respect to the demographic shock and the effects of technological change. These predictions can be checked against the historical performance of our three geographical areas: Byzantium, the Islamic world, and Western Europe. In Byzantium the plague (and the subsequent invasions by the Arabs) led to a contraction of the economy. The area under cultivation was reduced and as a result urban life underwent a decline as well. Per capita income, however, increased as the land-labor ratio rose, which made it possible to shoulder a higher military burden per capita. The increase in per capita income induced population growth from some time between the late eighth century and the early eleventh, a process that also received some aid from the more peaceful conditions prevailing on the political frontier. As predicted by our model, when the population began to grow back, bargaining power in the labor and land markets shifted in favor of the landowners, and this resulted in harsher contractual conditions for the peasantry. Land rents increased at the expense of wages. At the same time, political conditions deteriorated, with the military frontier being pushed in by the Seljuk Turks and the Fourth Crusade. This opened the door for the eventual collapse of the empire with the fall of Constantinople in 1453. Nowhere in this process does technological progress appear to have been present, so that the Byzantine story was simply one of population change and military operations within a shrinking territory and a concomitant loss of population to superior military force. The core territory of what would later become the Abode of Islam escaped the ravages of the plague. The cycle began by territorial expansion. The military frontier between the Byzantine and Sassanid empires was weakened by the loss of population due to the plague, and into this population vacuum the Arabs could move relatively easily, pushing both north and west into Byzantine territory in northern Africa and east into Iran all the way to the region of Sind in the west of India. The area of arable land was extended not by clearance but
194
Historical Applications of Heckscher-Ohlin Theory
by conquest. Territorial expansion was, however, accompanied by remarkable technological progress in agriculture. New crops stimulated the use of irrigation and land clearance, industrial processing of agricultural products, and also demographic growth and urban development. The population of the Islamic world would grow from some time between 700 and 750 to 1000 at a rate that was superior to that of Byzantium, but not one that could be called a ‘‘demographic revolution.’’ Such a revolution was not possible under the unfavorable conditions of the physical environment. The largely desertlike conditions in much of the territory of Islam could not support a higher population. To this can be added the unfortunate impact of certain social institutions and the devastating blow dealt to Islam by the Mongols. The last of our three cases, Western Europe, emerges by a wide margin as the winner in our imaginary population race. Of course, we can only speculate about the reasons for this, in the light of our model. One factor was technological change: the improvement of the plow, the substitution of the horse for the ox, the horse collar, the horseshoe, the three-field system of crop rotation, and the change in the crop mix, as well as the concomitant changes in the social system. More important, however, was what took place at the frontier of cultivation. In different ways the frontier was spectacularly extended. The dense forests had to yield to the sown. While in the case of the Islamic world the desert increasingly encroached on the sown, in Western Europe population growth induced by technical progress ensured the retreat of the wilderness. The Germans advanced towards the east across the Elbe in an early Drang nach Osten, and the Spaniards pushed back the Moors, in an effort that would carry over to overseas conquests beginning in 1492, the very year that the military frontier was closed in the Iberian Peninsula itself. The results of all these frontier movements were spectacular, far above the achievements by Byzantium and Islam. Cities grew, albeit not to the size of those of Byzantium and Islam, but grow they did, both in size and in numbers. Western Europe experienced the same cycle as the other two regions, first a downswing, then an upturn in population. The upturn was nothing short of spectacular in the Malthusian age that we are dealing with, to the point where at the end of the period symptoms of serious overpopulation began to appear. The advent of the Black Death drastically raised the land-labor ratio back to a much more favorable level. In a perverse sense it came as a deliverer, triggering a new demographic-economic cycle that would last
Demographic Shocks and the Factor Proportions Model
195
all the way up to the Industrial Revolution (Findlay and Lundahl 2002). Notes 1. The traditional date is October 732, but it could also be October 733 (Fouracre 2000:87). 2. Livi-Bacci (2001:27), quoting Biraben (1979:16), has similar figures: 200, 44 million; 600, 22 million; and 1340, 74 million. 3. The plague may not have been the only disease to gain a strong foothold in late antiquity. As McCormick (2001:38–41) has pointed out, both malaria and leprosy appear to have been on the rise in the Roman world at about the same time as when the plague of Justinian struck. 4. Let us assume that the introduction of the plow saves N man-days per hectare, and that the daily wage rate is w. The annual capital cost (the sum of depreciation and interest, which we assume to be given), of plowing S hectares is c ¼ SNw, and S ¼ ðc=wÞð1=NÞ defines the lowest (‘‘threshold’’) farm size that makes it profitable to introduce the plow. 5. The Domesday Book was put together some time between 1108 and 1109 or between 1111 and 1113 (Poole 1955:1 n.).
References Bartlett, R., and A. Mackay. 1989. Medieval Frontier Societies. Oxford: Clarendon Press. Biraben, J.-N. 1979. ‘‘Essai sur l’e´volution du nombre des hommes.’’ Population 34:13–25. Biraben, J.-N., and J. LeGoff. 1969. ‘‘La peste dans le haut moyen aˆge.’’ Annales: E´conomie, Socie´te´, Civilisations 24:1484–1510. Bloch, M. 1962. Feudal Society. 2nd ed. London: Routledge and Kegan Paul. Boserup, E. 1965. The Conditions of Agricultural Growth: The Economics of Agrarian Change Under Population Pressure. London: George Allen and Unwin. Bray, R. S. 2000. Armies of Pestilence: The Impact of Disease on History. New York: Barnes and Noble. Cameron, A. 1993. The Mediterranean World in Late Antiquity. London: Routledge. Collins, R. 1995. Early Medieval Spain: Unity in Diversity, 400–1000. London: Macmillan. Diamond, J. 1997. Guns, Germs and Steel: The Fates of Human Societies. New York: Norton. Dols, M. W. 1974. ‘‘Plague in Early Islamic History.’’ Journal of the American Orientalist Society 94:371–383. Duby, G. 1974. The Early Growth of the European Economy: Warriors and Peasants from the Seventh to the Twelfth Century. London: Weidenfeld and Nicolson. ———. 1980. The Three Orders. Chicago: University of Chicago Press.
196
Historical Applications of Heckscher-Ohlin Theory
Findlay, R. 1993. ‘‘International Trade and Factor Mobility with an Endogenous Land Frontier: The General Equilibrium Consequences of Christopher Colombus.’’ In W. J. Ethier et al., eds., Theory, Policy and Dynamics in International Trade, 38–54. Cambridge: Cambridge University Press. ———. 1996. ‘‘Towards a Model of Territorial Expansion and the Limits of Empire.’’ In M. R. Garfinkel and S. Skaperdas, eds., The Political Economy of Conflict and Cooperation, 41–56. Cambridge: Cambridge University Press. ———. 1998. ‘‘A Plea for Trade Theory in Economic History.’’ Economic and Social Review 29:313–321. Findlay, R., and M. Lundahl. 2002. ‘‘Toward a Factor Proportions Approach to Economic History: Population, Precious Metals, and Prices from the Black Death to the Price Revolution.’’ In R. Findlay, L. Jonung, and M. Lundahl, eds., Bertil Ohlin: A Centennial Celebration (1899–1999), 495–528. Cambridge, MA: MIT Press. ———. In press. ‘‘The First Globalization Episode: The Creation of the Mongol Empire, or the Economics of Chinggis Khan.’’ In G. Therborn and H. H. Khondkar, eds., Asia and Europe in Globalization: Continents, Regions, and Nations. Leiden: Brill. Fossier, R. 1999. ‘‘Rural Economy and Country Life.’’ In T. Reuter, ed., The New Cambridge Medieval History, vol. III c. 900–c. 1024. Cambridge: Cambridge University Press. Fouracre, P. 2000. The Age of Charles Martel. Harlow: Longman. Gregory of Tours. 1974. The History of the Franks. Trans. L. Thorpe. Harmondsworth: Penguin. Grigg, D. B. 1980. Population Growth and Agrarian Change: An Historical Perspective. Cambridge: Cambridge University Press. Harrison, D. 1999. Krigarnas och helgonens tid: Va¨steuropas historia 400–800 e Kr. Stockholm: Prisma. Harvey, A. 1989. Economic Expansion in the Byzantine Empire 900–1200. Cambridge: Cambridge University Press. Heckscher, E. F. 1929. ‘‘A Plea for Theory in Economic History.’’ Economic History (supplement to Economic Journal) 1:525–534. Henriksson, R. G. H., and M. Lundahl. 2003. ‘‘Eli Heckscher, ekonomisk teori och ekonomisk historia.’’ In Janusansiktet Eli Heckscher: Nationalekonom och ekonomisk historiker. Texter i urval av Rolf G. H. Henriksson och Mats Lundahl. Stockholm: Timbro. Hitti, P. K. 1970. History of the Arabs: From the Earliest Times to the Present. 10th ed. London: Macmillan. Issawi, C. 1981. ‘‘The Area and Population of the Arab Empire: An Essay in Speculation.’’ In A. L. Udovitch, ed., The Islamic Middle East 700–1900. Princeton, NJ: Darwin Press. Jordan, W. C. 1996. The Great Famine: Northern Europe in the Early Fourteenth Century. Princeton, NJ: Princeton University Press. Kazdahn, A. P., and A. W. Epstein. 1985. Change in Byzantine Culture in the Eleventh and Twelfth Centuries. Berkeley: University of California Press. Khaldun, Ibn. 1958. The Muqaddimah: An Introduction to History. 3 vols. Trans. F. Rosenthal. New York: Pantheon Books.
Demographic Shocks and the Factor Proportions Model
197
Lewis, A. R. 1958. ‘‘The Closing of the Medieval Frontier 1250–1350.’’ Speculum 33:475–483. Livi-Bacci, M. 2001. A Concise History of World Population. 3rd ed. Malden, MA: Blackwell. Lopez, R. S. 1976. The Commercial Revolution of the Middle Ages, 950–1350. Cambridge: Cambridge University Press. Lundahl, M. 1979. Peasants and Poverty: A Study of Haiti. London: Croom Helm. Mango, C. 1994. Byzantium: The Empire of the New Rome. London: Phoenix. McCormick, M. 2001. Origins of the European Economy: Communications and Commerce, A.D. 300–900. Cambridge: Cambridge University Press. McEvedy, C., and R. Jones. 1980. Atlas of World Population History. Harmondsworth: Penguin. Nicol, D. M. 1988. Byzantium and Venice: A Study in Diplomatic and Cultural Relations. Cambridge: Cambridge University Press. North, D. C., and R. P. Thomas. 1973. The Rise of the Western World: A New Economic History. Cambridge: Cambridge University Press. Norwich, J. J. 1991. The Normans in Sicily. Harmondsworth: Penguin. ———. 1997. Byzantium: The Decline and Fall. New York: Knopf. Ocan˜a Jime´nez, M. 1975. ‘‘Co´rdoba musulmana.’’ In Juan Bernier Luque et al., Co´rdoba: colonia romana, corte de los califas, luz de occidente, 25–48. Madrid: Editorial Everest. Ostrogorsky, G. 1966. ‘‘Agrarian Conditions in the Byzantine Empire in the Middle Ages.’’ In M. M. Postan, ed., The Cambridge Economic History of Europe, vol. I, The Agrarian Life of the Middle Ages, 205–234. 2nd ed. Cambridge: Cambridge University Press. Planhol, X. de. 1959. The World of Islam. Ithaca, NY: Cornell University Press. Poole, A. L. 1955. From Domesday Book to Magna Carta 1087–1216. Oxford: Clarendon Press. Postan, M. M. 1966. ‘‘Medieval Agrarian Society in Its Prime: England.’’ In M. M. Postan, ed., The Cambridge Economic History of Europe, vol. I, The Agrarian Life of the Middle Ages, 549–632. 2nd ed. Cambridge: Cambridge University Press. Queller, D. E., and T. F. Madden. 1997. The Fourth Crusade: The Conquest of Constantinople. 2nd ed. Philadelphia: University of Pennsylvania Press. Riley-Smith, J. 1987. The Crusades: A Short History. New Haven, CT: Yale University Press. Runciman, S. 1951. A History of the Crusades. 3 vols. Cambridge: Cambridge University Press. Russell, J. C. 1958. ‘‘Late Ancient and Medieval Population.’’ Transactions of the American Philosophical Society 48:3–152. ———. 1968. ‘‘That Earlier Plague.’’ Demography 5:174–184. Saunders, J. J. 1965. A History of Medieval Islam. London: Routledge and Kegan Paul. Sawyer, P. H., and R. Hilton. 1963. ‘‘Technical Determinism: The Stirrup and the Plough.’’ Past and Present 24:90–100.
198
Historical Applications of Heckscher-Ohlin Theory
Treadgold, W. 1997. A History of the Byzantine State and Society. Stanford, CA: Stanford University Press. ———. 2001. A Concise History of Byzantium. Houndmills: Palgrave. ———. 2002. ‘‘The Struggle for Survival (641–780).’’ In C. Mango, ed., The Oxford History of Byzantium, 129–150. Oxford: Oxford University Press. Turner, F. J. 1986. The Frontier in American History. Tucson: University of Arizona Press. Van Werveke, H. 1963. ‘‘The Rise of Towns.’’ In M. M. Postan, E. E. Rich, and E. Miller, eds., The Cambridge Economic History of Europe, vol. III, Economic Organization and Policies in the Middle Ages, 3–41. Cambridge: Cambridge University Press. Watson, A. M. 1983. Agricultural Innovation in the Early Islamic World. Cambridge: Cambridge University Press. White, L. 1962. Medieval Technology and Social Change. Oxford: Oxford University Press. ———. 1978. Medieval Religion and Technology: Collected Essays. Berkeley: University of California Press. Wolf, E. 1966. Peasants. Englewood Cliffs, NJ: Prentice-Hall. Zinsser, H. 1935. Rats, Lice and History. London: Routledge and Sons.
8
Explaining World Tariffs, 1870–1938: StolperSamuelson, Strategic Tariffs, and State Revenues Jeffrey G. Williamson
What determines tariff policy? It can’t be conventional economics, since every mainstream economist agrees that free trade is a good thing (Smith 1776; Mill 1909; Krugman 1996; Bhagwati 2000). Yet, the politics of free trade have been surrounded by controversy ever since Alexander Hamilton tried shoving his protectionist policies down the throats of a new United States federal congress after 1789, and since Robert Peel ruined his political career by shoving free trade down the throats of the British Parliament in 1846. Political leaders have never been solely, or even largely, interested in maximizing national income, let alone maximizing world income. Rather, their main goal has always been ‘‘to get a larger slice [of the pie] for their supporters’’ (McGillivray et al. 2001:2). Protection and free trade have always been for sale in the political market place (Grossman and Helpman 1994), but having said so doesn’t make the question—What determines tariff policy?—much easier to answer. After all, nations will adopt different tariff policies to the extent that there are different economic interests lobbying for those policies, to the extent that the economic environment impacting on those interests is different, and to the extent that different political institutions dictate which economic interests have the most votes. Thus, to explain tariff policy, we need to understand the underlying economic, political and institutional fundamentals at work. As a recent book collaboration by four political scientists has pointed out so effectively (McGillivray et al. 2001:3–16), there are three ways that endogenous tariff theory has confronted fact as it has sought to uncover the fundamentals: first, by comparing tariffs by industries within countries; second, by comparing countries at various points in time; and third, by exploring tariffs over time. It is fair to say that the vast majority of the empirical work on endogenous tariffs has elected the first route—within-country variance across industries, and most of it is on
200
Historical Applications of Heckscher-Ohlin Theory
the post–World War II United States (e.g., Magee, Brock, and Young 1989; but for pre–World War II United States, see Pincus 1977 and Marvel and Ray 1983, 1987). The second route—variance across countries—has been exploited less intensively (e.g., Kindleberger 1951; Conybeare 1983; Magee, Brock, and Young 1989). The third route— variance over time—has been exploited the least (e.g., Magee, Brock, and Young 1989). A fourth route has been added recently—panel data (Coatsworth and Williamson 2004a; Blattman, Clemens, and Williamson 2002; Clemens and Williamson 2002). This essay will exploit the second, third, and fourth routes by first exploring quantitatively annual tariff rate data for thirty-five countries between 1870 and 1938 (accounting for 82.8 percent of the world’s 1900 population). The next sections will review the pre–World War II tariff evidence, identifying the tariff facts most needing explanation; explore the familiar Stolper-Samuelson corollary and its recent extensions, showing why it is so important to find out whether it has been the central force driving tariffs in the past; do the same for the infant industry argument; and lay out the contending determinants of tariff policy and explore their role empirically. Here, world tariffs are treated as country and time fixed effects in the seven decades before 1938, a period that contains both the first global century up to World War I and the interwar autarkic disaster that followed. It appears that tariff policy before World War II was driven primarily by Stolper-Samuelson forces, revenue needs and strategic tariff behavior, not by infant industry. World Tariffs 1870–1938: The Facts This essay uses the computed average tariff rate1 to explore the policy experience of thirty-five countries the world around between the 1860s and World War II: the United States; three members of the European industrial core (France, Germany, United Kingdom); three non-Latin European offshoots (Australia, Canada, New Zealand); ten from the industrially lagging European periphery (Austria-Hungary, Denmark, Greece, Italy, Norway, Portugal, Russia, Serbia, Spain, Sweden); ten from Asia and the Mideast (Burma, Ceylon, China, Egypt, India, Indonesia, Japan, the Philippines, Siam, Turkey); and eight from Latin America (Argentina, Brazil, Chile, Colombia, Cuba, Mexico, Peru, Uruguay). Figure 8.1 plots average world tariffs from the 1860s to the 1990s, and figure 8.2 plots it up to 1938 for some regional clubs.2 There are six regions plotted in figure 8.2—the United States, the European
Explaining World Tariffs, 1870–1938
Figure 8.1 Unweighted world average own tariff, 35 countries Source: Blattman, Clemens, and Williamson (2002, figure 1)
Figure 8.2 Unweighted world average of regional tariffs before World War II Source: Coatsworth and Williamson (2002, figure 2)
201
202
Historical Applications of Heckscher-Ohlin Theory
industrial core, the European periphery, the European non-Latin offshoots, Asia, and Latin America—the country members of which have just been identified. Note first the powerful role played by inflations and deflations at key points in the past. Import duties were typically specific until modern times, quoted as pesos per bale, dollars per yard, or yen per ton. Under a regime of specific duties, abrupt changes in price levels can change import values in the denominator, but not the legislated duty in the numerator, thus producing big percentage point changes in equivalent ad valorem tariff rates. The impact of inflation during World War I was quite spectacular, and it had nothing to do with policy. Thus, tariff rates in all six regions fell sharply between 1914 and 1919,3 and part of the rise in tariffs immediately after the war was also due to postwar deflation and the partial resumption of prewar price levels. The price deflation after 1929 was even more spectacular, and it too served to raise tariff rates at least on duties that were still specific (import values now declining). While the specific-duty effect certainly played a role worldwide at these critical points, it was not an important factor in accounting for differences between countries or for long run trends. Second, the well-known surge to world protection in the 1920s and 1930s is certainly revealed in figure 8.1. What is less well known, however, is the pronounced protectionist drift worldwide between 1865 and about 1900. And what looks in figure 8.1 like a modest pre–World War I antiglobalization backlash—a retreat from the liberal proglobal trade positions in midcentury (Williamson 1998, 2002)—is far more dramatic when the world averages are disaggregated in figure 8.2. Indeed, there is a very pronounced rise in tariffs across Latin America, across the non-Latin European offshoots (the United States being the major exception) and across the European periphery. This steep rise up to the 1890s in the periphery’s tariff rates far exceeds that of the European core, a notable fact given that almost nothing has been written on this antiglobal tariff trend in the periphery. Third, note the enormous variance in levels of protection between the regional averages. The richer New World European offshoots had levels of protection almost three times that of the European core around the turn of the last century. When the United States is shifted to the rich European offshoot club, the ratio of European offshoot tariffs to that of the core is more than three to one. To take another example,
Explaining World Tariffs, 1870–1938
203
in 1925 the European periphery had tariffs about two and a half times higher than those in the European part of the industrial core. To take yet another example, in 1885 the poor but independent parts of Latin America (Brazil, Colombia, Mexico, and Peru) had tariffs almost five times higher than those in the poor and dependent parts of Asia (Burma, Ceylon, China, Egypt, India, Indonesia, and the Philippines), while the poor but independent parts of Asia (Siam, Turkey, and Japan) had tariff rates about the same as the poor but dependent parts of Asia. Of course, colonial status, lack of autonomy, and ‘‘unequal treaties’’ all played an important role in Asia, and we will want to control for that fact in what follows. Fourth, there was great variance within these regional clubs. In 1905, tariffs in Uruguay (the most protectionist land-abundant and laborscarce country) were about two and a half times those in Canada (the least protectionist land-abundant and labor-scarce country). In the same year, tariffs in Brazil and Colombia (the most protectionist poor but autonomous countries in Latin America) were almost ten times those in China and India (the least protectionist poor and nonautonomous countries in Asia). The same high-low range appeared within the industrial core (the United States five times the United Kingdom) and the European periphery (Russia six times Austria-Hungary). Between 1919 and 1938, the tariff variance between countries was about the same as tariff variance over time, but between 1865 and 1914, the tariff variance between countries was more than twice that of the tariff variance over time. Thus, explaining differences in tariff policy between countries is at least as challenging as explaining changes in tariff policy over the eight decades after the 1860s, perhaps more so. The empirical analysis later in this essay will treat countries as the unit of observation, but for a moment let us linger a little longer on the regional clubs. Prior to World War I, tariffs were much higher in the rich European offshoots than anywhere else. Furthermore, and as I have already mentioned, they would have been even higher had I allocated to this club one of the most protectionist, the United States (which is allocated instead to the core).4 The European members of the industrial core (France, Germany, the United Kingdom) had the lowest tariffs, although the United States serves to raise the club average. Most members of the poor periphery in Asia were colonies or quasi-colonies of the industrial core (Burma, Ceylon, Egypt, India, Indonesia, the Philippines), or were forced to sign free trade agreements (‘‘unequal
204
Historical Applications of Heckscher-Ohlin Theory
treaties’’) with the core since the latter had naval guns trained on their potential trading partners (China, Japan), or viewed nearby gunboats as a sufficient threat to go open on their own (Siam). Thus, tariff rates in Asia were pretty much like those of the core early on, but they started drifting toward protection after the 1880s, long before the post– World War II independence movement. It should also be stressed that colonial status did not necessarily imply lack of local influence on tariff policy. There are five colonies in our sample from Asia: Burma, Ceylon, India, Indonesia, and the Philippines, although foreign influence was strong enough (including occupation) to make Egypt behave like a colony. A previous paper (Clemens and Williamson 2002) has shown that while colonial tariff policy did indeed mimic that of their masters, local conditions mattered as well. Thus, I retain the full sample of thirty-five, although I will take care to control for colonial status and tariff autonomy. In any case, while Asia had the lowest tariffs in 1865, they were approaching that of the protectionist rich European offshoots by 1914. The European periphery leaped to high levels of protection after the 1870s, with Russia leading the way. There is plenty of evidence of rising world protection before World War I (the unweighted average in the full sample rising from about 12 percent in 1865 to about 17.5 percent in 1900), but the much-studied European continental backlash plotted in figure 8.2 looks pretty modest compared with the rest of the world. Indeed, the pre-1914 global backlash took place mainly in Latin America, the European offshoots (excluding the United States, which retreated from its enormous Civil War tariffs), and the European periphery. There are some surprises in these tariff data that have not been given much notice by previous scholars. For example, the traditional literature has made much of the tariff backlash on the continent to the ‘‘grain invasion’’ after the 1870s (Kindleberger 1951; Bairoch 1989; O’Rourke 1997): between the 1870s and the 1890s, average tariff rates rose by 5.7 percentage points in France (to 10.1 percent) and 5.3 percentage points in Germany (to 9.1 percent). However, this antiliberal move to higher tariffs by the leading economies on the continent is repeated in the European periphery (up 4.2 percentage points to 16.8 percent) and in our four poor Latin American countries (up 6.9 percentage points to 34 percent), regions where, one assumes, a ‘‘manufactures invasion’’ must have been the motivating event. The traditional literature also teaches
Explaining World Tariffs, 1870–1938
205
that the Latin American reluctance to go open in the late twentieth century was the product of the Great Depression and the import substitution industrial (ISI) strategies that arose from it (Diaz-Alejandro 1984; Corbo 1992). Yet, Latin America already had by far the highest tariffs in the world by the mid- to late nineteenth century (Coatsworth and Williamson 2004a, 2004b). Thus, whatever explanation is offered for the Latin American commitment to high tariffs, it must search for origins well before the Great Depression. Finally, it is not true that Asia waited for post–World War II independence to switch to protectionist policies. I have already noted that there was an upward surge in tariff rates in Asia after the 1880s and early 1890s, illustrated best by Burma, India, the Philippines, Siam, and Turkey. With the exception of Egypt and Japan, all of the Asian countries underwent a surge to high tariffs in the 1930s, and most of these countries stuck with these higher tariffs into the 1940s and the modern era. How much of the surge in Asian tariffs from the 1880s to the 1930s was due to a weakening colonial grip and to the expiration of ‘‘unequal treaties’’ signed decades earlier, both of which would have given the region the increasing autonomy to set higher tariffs according to local political economy forces? Tariffs took two big leaps upward in the interwar decades, and these took place world wide. The first leap was in the 1920s, which might be interpreted as a return to high prewar tariff rates. The second was in the 1930s, with the well-known and aggressive beggar-my-neighbor policies. The biggest interwar tariff hikes in the industrial core were initiated by Germany and the United Kingdom, but France and the United States were not far behind. Indeed, the rise in tariff rates was so pronounced in the core, that the big pre-1914 spread between the hightariff autonomous periphery and the low-tariff industrial core evaporated. Still, tariffs rose in most of the European periphery and almost everywhere in Latin America, Asia and the Middle East. To give some sense of how large the rise in tariff barriers was around an Asian periphery dominated by allegedly passive and free-trading colonies, the tariff rate rose in India by 22 percentage points between 1920 and 1939, in Egypt it rose by 36.7 percentage points between 1920 and 1939, in Siam it rose by 26.9 percentage points between 1918 and 1936, and in Turkey it rose by 34.1 percentage points between 1923 and 1937. So, what determined who protected and when in the century and a half before World War II?
206
Historical Applications of Heckscher-Ohlin Theory
Was It Stolper-Samuelson or Something Else, and Why Does It Matter? Was the rise in tariffs and/or high tariffs before World War I driven by some antiglobal reaction, that is, by some backlash? Until the race towards autarky in the 1930s, the free traders were members of the industrial core, their colonies, or those who their gunboats had intimidated to open up. The rest had erected high tariff walls. Was the autonomous periphery exhibiting global backlash? In the three decades or so following 1865, the rise in tariff rates was ubiquitous worldwide. Was this upsurge a policy backlash response to the spectacular fall in transport costs which was serving to integrate world commodity markets and to blow the winds of international competition down the necks of import-competing industries which geography had protected before? It is essential to get answers to these questions if the modern debate about the future of globalization is to be properly informed by history. Simply to show high and/or rising tariffs is not enough. Did globalization backlash account for it? The most elegant backlash explanation has its roots in Stockholm. Eli Heckscher and Bertil Ohlin told us how endowments could account for trade patterns, factor abundance dictating competitiveness in world markets and what would be exported by whom. In addition, Heckscher (1991) showed how foreign trade effects the distribution of income, but most economists had to wait for Wolfgang Stolper and Paul Samuelson (1941) to elaborate the corollary in English, namely that the scarce factor should favor protection and the abundant factor should favor free trade. A decade or so ago, (Heckscher-) StolperSamuelson thinking was used with great skill by Ronald Rogowski (1989) who applied it to country trade policy the world around from 1840 to the present. There are two limitations to the way Rogowski uses the Stolper-Samuelson corollary, however. The first limitation is that the corollary only tells us who votes for what, not who wins the voting. Since the landed elite dominated voting in land-scarce Europe,5 the import-competing sectors got the protection from foreign grains that the landed elite wanted. Trade theorists have, in fact, offered an explicit rule (the ‘‘endowment effect’’) whereby the ‘‘equilibrium tariff increases with the square root of the ratio of the country’s scarce factor to its abundant factor’’ (Magee, Brock, and Young 1989:25). Alternatively, as the scarce factor shrinks in relative size, its power at the polls shrinks too. However, what happens to such endowment rules when
Explaining World Tariffs, 1870–1938
207
the scarce factor does not have the vote, as was true of labor throughout most of the world before the 1930s?6 Did labor get the protection of import-competing manufacturing that it should have wanted in labor-scarce Latin America, in the labor-scarce English-speaking new world, and labor-scarce southeast Asia? The second limitation is that Rogowski uses the corollary to speak only to levels of protection rather than to changes in protection. We want to know whether a rise in protection can be attributed to globalization backlash and to compensation of the damaged scarce factor, so a dynamic version of StolperSamuelson is more relevant. While the Ricardo-Viner-Cairnes specific-factor model yields results similar to the Stolper-Samuelson model—import-competing industries favor protection, (Heckscher-) Stolper-Samuelson thinking is probably more effective for long-run analysis like that contained in this essay. More generally, when the import-competing sector is damaged by an adverse price shock (an improvement in the country’s terms of trade) induced by world market events or by declining seaborne transport costs that reduce import and raise export prices, is there always a ‘‘compensation effect’’ that drives up tariffs? The answer will depend largely on whether the factors in the slumping sector can escape to the booming sector. Stolper-Samuelson has a far better chance of explaining nineteenth-century tariff policy when, after all, most trade was in primary products and (immobile) specific factors played a big role. It has a far poorer chance of explaining modern tariff policy when trade is dominated by manufactures and most factors—labor, skills and capital—are mobile.7 There is no shortage of elegant backlash models. It’s the evidence brought to bear that’s scarce. Was It Infant Industry or Something Else, and Why Does It Matter? It has always been believed that a second powerful motivation for high tariffs on imported manufactures in the early-industrial periphery is development policy. Central authorities were persuaded for much of the twentieth century that industrialization was the only vehicle for development and that protection fostered that process. Indeed, they have often cited nineteenth-century experience to help support these claims. I will call this motivation the infant industry argument for short, with the understanding that it includes development and industrial policy.
208
Historical Applications of Heckscher-Ohlin Theory
Does protection help or hinder growth? It should be useful to answer this question first to see whether policy makers in the autonomous parts of the periphery could have used such evidence to support their protectionist policies in the century before the 1930s. Of course, policy makers of that time didn’t have the models, methods, and evidence that we can exploit today, but they certainly would have had the intuition. Were we asking this question about the late twentieth century, then the evidence would strongly support the position that protection hindered growth. But what about the nineteenth century? Did protection foster8 growth in the pre–World War I periphery where those tariff rates were so high? Policy makers in those parts of the periphery which had tariff autonomy were certainly aware of the pro-protectionist infant-industry argument offered for the German Zollverein by Frederich List and for the United States customs union by Alexander Hamilton. This was certainly true of late-nineteenth-century Latin America (Bulmer-Thomas 1994:140). However, it is important to stress ‘‘late’’ in the previous sentence since the use of protection specifically and consciously to foster industry does not occur in Mexico until the early 1890s, Brazil and Chile a little later in the 1890s, and Colombia in the early 1900s (Coatsworth and Williamson 2004a, 2004b). So, the qualitative evidence suggests that domestic industry protection becomes a significant motivation for Latin American tariffs only near the turn of the previous century. It turns out that there is absolutely no pre–World War I quantitative evidence which would have supported infant industry arguments for Latin America either: high tariffs were correlated with slow growth, just like the late twentieth century (Coatsworth and Williamson 2004a). We must look elsewhere for plausible explanations for the exceptionally high (and often rising) tariffs in the autonomous periphery in the century before the Great Depression. One of the alternative explanations that I will explore in the next section involves the revenue needs of central governments. As a signal of things to come, I simply note here that the causation probably went the other way round in the autonomous periphery. That is, countries achieving rapid GDP per capita growth also had underwent faster growth in imports and in other parts of the tax base, thus reducing the need for high tariff rates. And countries suffering slow growth would have had to keep tariff rates high to ensure adequate revenues. Why is it important to find no evidence supporting a protectionfosters-growth correlation in the periphery before the interwar de-
Explaining World Tariffs, 1870–1938
209
cades? The answer, of course, is that such evidence would add more support to the view that those high and rising tariffs in the periphery represented a ‘‘globalization backlash’’ as it was ‘‘flooded with manufactures’’ from the industrial core.9 The periphery was flooded with ever-cheaper manufactures, as the natural barriers of geography fell in response to the railroad and steamship, and as industrial Europe and North America underwent impressive productivity advance in manufacturing. If the periphery was in fact hoping to stimulate industrial development by protection, tariffs would have had to rise higher and higher to offset the continued fall in the landed price of imported manufactures. The Political Economy of Tariffs: Some Preliminaries Tariffs for Revenue Were revenues a strong motive for high tariffs? If so, were those high pre–World War I tariffs in Latin America and the European periphery really all that the market could bear? As Douglas Irwin (1998a:8–12) has recently stressed for the United States, the revenue-maximizing tariff hinges crucially on the price elasticity of import demand. Tariff revenue can be expressed as R ¼ tpM, where R is revenue, t is the average ad valorem tariff rate, p is the average import price, and M is import volume. Totally differentiating with respect to t, and assuming that the typical nineteenth century country in the periphery was a price taker for manufacturing imports, yields dR=dt ¼ pM þ ðtpÞ dM=dt. The revenue-maximizing tariff rate, t , is found by setting dR=dt ¼ 0—the peak of some Laffer curve—in which case t ¼ 1=ð1 þ hÞ, where h is the price elasticity of demand for imports. Irwin (1998a:14) estimates the price elasticity to have been about 2:6 for the United States between 1869 and 1913. Since the import mix for countries around the periphery was similar to that of the United States, assuming the price elasticity for the former around 3 can’t be too far off the mark. Under those assumptions, the revenue-maximizing tariff in the periphery would have been very high indeed, about 50 percent. Suppose some government in the periphery—riding an export boom—had in mind some target revenue share in GDP ðR=Y ¼ rÞ and could not rely on foreign capital inflows to balance the current account (so pM ¼ X), then r ¼ tpM=Y ¼ tX=Y. Clearly, if foreign exchange earnings from exports (and thus imports) were booming (an event which could be caused by a terms of trade boom, denoted here by a
210
Historical Applications of Heckscher-Ohlin Theory
fall in the relative price of imports, p, or by a supply-side expansion which increased export quantities, X), then the target revenue share could have been achieved at lower tariff rates, t. The bigger the export boom, the higher the export share, the bigger the import share, and the lower the necessary tariff rate. So, did independent governments in Latin America, the European periphery, and Asia act as if they were meeting revenue targets? Ceteris paribus, did they lower tariff rates during world primary product booms when export shares were high and rising, and did they raise them during world primary product slumps? Of course, countries in the periphery that were successful in getting external finance from the European core would have had less reason to use high tariffs to augment revenues in the short run and medium term. Since world capital markets became increasingly well integrated up to 1913 (Obstfeld and Taylor 2003), high tariffs that were necessary in 1865 would no longer have been necessary in 1913 if ‘‘revenue smoothing’’ was a key motivation. However, there may have been plenty of motivation to raise them again when world capital markets fell apart in the interwar years. Furthermore, countries that developed internal (and less distortionary) tax sources would have had less need for high tariffs, an event that started in the late-nineteenth-century industrial core, accelerating during the interwar rise of the welfare state (Lindert 1994). Such developments lagged behind in the periphery, however. The (Heckscher-) Stolper-Samuelson Theorem and Scarce Factor Compensation The Stolper-Samuelson theorem tells us that protection benefits owners of factors in which that society is poorly endowed. According to this kind of thinking, Latin American capitalists should have been looking to form protectionist coalitions as soon as the belle e´poque began to threaten them with freer trade. In most cases, they did not have to look far, either because they managed to dominate oligarchic regimes that excluded other interests, or because they readily found coalition partners willing to help, or both (Rogowski 1989; Coatsworth and Williamson 2004a). Why no scarce labor in the Latin American tale? Growth, peace, and political stability after 1870 did not necessarily produce democratic inclusion in Latin America. Most countries in the region limited the
Explaining World Tariffs, 1870–1938
211
franchise to a small minority of adult men until well into the twentieth century. Literacy and wealth requirements excluded, as we have seen, most potential voters in virtually every country (Engerman, Haber, and Sokoloff 2000). Thus, the late nineteenth century tended to produce oligarchic governments in which urban capitalists—linked to external trade and finance—played a dominant role. In countries that specialized in exporting agricultural products, free-trading landowners formed the second dominant part of the governing oligarchy. Freetrading mineral export interests usually had less direct leverage in governmental decision making, despite the size and significance of their investments. Thus, unambiguous protectionist outcomes would hardly have been predicted for every Latin American country. To the extent that Stolper-Samuelson thinking is useful in accounting for the variance in tariff rates the world around before World War II, we would expect plenty of regional differences, as Rogowski has argued. After all, very different endowments and political participation characterized various parts of the periphery. The land-abundant English-speaking New World countries were places where scarce labor had a powerful political voice to lobby for protection, joining scarce capital. The European periphery had scarce land and capital lobbying for protection, while the voices of free-trading labor were suppressed. Southeast Asia had scarce labor and capital, but with political participation limited to free-trading landed interests. The rest of Asia was pretty much land and capital scarce, but free-trading labor had little or no political voice. The important point here is that the StolperSamuelson theorem tells us who should vote for free trade and who should vote for protection, but it does not tell us who gets the most votes. Productivity Advance Abroad, Deindustrialization Fears and Scarce Factor Compensation Were high and rising tariffs in the periphery generated by deindustrialization fears and/or Stolper-Samuelson compensation of scarce factors at home in response to falling import prices? Three things are essential to the survival of domestic industry (using scarce factors): low costs of inputs—like labor, power, and raw materials; high productivity in the use of those inputs; and high market prices of output. Policy makers in the periphery could not do much about the first two,10 but they could do a great deal about the third by
212
Historical Applications of Heckscher-Ohlin Theory
pushing up tariff barriers, excluding foreign imports and thus raising the domestic price of manufactures relative to other products produced for home or foreign markets. When industrial productivity advance in the core was fast, world market prices of manufactures would decline relative to other products, and foreign firms would be increasingly competitive in local periphery markets. Thus, policy makers in the periphery who favored industry, and/or the scarce factors used there, would have had reason to raise tariffs in response to any sharp decline in the relative price of manufactures, especially relative to prices of the primary products the periphery exported to Europe. In short, if the periphery had deindustrialization fears, or wanted to compensate the damaged scarce factors, it would have raised tariffs in response to falling prices of manufactures in world markets, that is in response to a rise in the world price of primary products, and thus to an improvement in the periphery’s terms of trade. Evaporating Geographic Barriers, Tariff-Transport Cost Trade-offs, and Scarce Factor Compensation High transport costs on goods imported from one’s trading partner are just as protective as high tariffs. When new transport technologies induce a dramatic fall in freight costs, the winds of competition thus created give powerful incentives to import competing industries (and scarce factors) to lobby for more protection. Since there certainly was a transport revolution across the nineteenth century (O’Rourke and Williamson 1999; Mohammed and Williamson 2004), there was plenty of incentive for manufacturing interests in the periphery and agricultural interests in the core to lobby for protection as the natural barriers afforded by transport costs melted away. This connection was confirmed long ago for the ‘‘invasion of grains’’ into Europe from the United States, the rest of the new world, and Russia. But what about the ‘‘invasion of manufactures’’ into the periphery from industrial Europe? The transport revolution took many forms, but three mattered most: a decline in overseas tramp freight rates; the appearance of major canals such as the Suez and the Panama; and the penetration of railroads into interior markets. Tramp freight rates fell everywhere, but mainly on routes carrying high bulk intermediates and foodstuffs to Europe, much less on routes carrying low bulk manufactures to the periphery. Meanwhile, railroads penetrated everywhere, and this fact
Explaining World Tariffs, 1870–1938
213
might have been especially relevant for tariff policy where markets were mainly located in the interior. If railroads exposed previously isolated interior local manufacturing to increased foreign competition, those interests should have lobbied for more protection, and railroad penetration of the interior was especially important in Latin America, Eastern Europe, and even India. Strategic Trade Policy, the Terms of Trade, and Tariffs A well-developed theoretical literature on strategic trade policy predicts that nations have an incentive to inflate their own terms of trade by raising tariffs, unless, of course, trading partners agree to mutual concessions (Dixit 1987; Bagwell and Staiger 2002). According to this kind of thinking, a country’s own tariffs will depend at least in part upon the country’s external tariff environment. Elsewhere, a principaltrading-partners’-tariff index has been calculated for our thirty-five countries (Blattman, Clemens, and Williamson 2002) and the index is revealing. In the two decades before World War I, every region except the industrial core and Latin America faced much lower tariff rates in their main export markets than they themselves erected against competitors in their own markets. The explanation, of course, is that the main export markets were located in the European core, where tariffs were much lower. Thus, most of the periphery faced much lower tariffs than did the core, although this was not true of Latin America for whom the protectionist United States was such an important market. During the interwar period there was convergence: every regional club faced very similar and high tariff rates in export markets, but those rates facing the periphery were rising very steeply as the core made that big policy switch from free trade to protection. It might pay to repeat that Latin America, for example, faced far higher tariffs than anyone else since they traded with the heavily protected United States. So, did this ‘‘hostile’’ policy environment abroad trigger a like response at home? While the strategic trade thesis holds promise in helping account for higher tariffs in Latin America and in that part of the European periphery trading with more-protectionist France and Germany, it holds less promise for that part of the European periphery whose exports were sent to free-trading United Kingdom. Indeed, between 1900 and World War I a decline in partner tariffs took place everywhere in the periphery except in the European periphery, suggesting a leader-follower reaction that varied
214
Historical Applications of Heckscher-Ohlin Theory
across the periphery depending on who the dominant trading partner was, for instance, an ultraprotectionist United States lowering tariffs, a moderately protectionist France and Germany raising tariffs, or a freetrade Britain standing pat (Blattman, Clemens, and Williamson 2002). Controls: Price Instability and the Specific-Duty Effect Inflations and deflations have had a powerful influence on average tariff rates. Recall that import duties were typically specific until modern times, quoted as pesos per bale, yen per yard, or dollars per bag. Under specific duty regimes, abrupt changes in price levels change import values in the denominator, but not the legislated duty in the numerator, thus producing big equivalent ad valorem or percentage rate changes. This specific-duty effect implies, of course, that debating the tariff structure is politically expensive, and thus is only infrequently changed by new legislation. The specific-duty effect has been explored most fully for the United States (Crucini 1994; Irwin 1998b:1017), but also for Mexico (Marquez 2002:307), and, more generally, for Latin America (Coatsworth and Williamson 2004a). The specific-duty effect has not, however, been explored at a global level. Nor does the literature tell us why specific duties seem to be much more common in young, nonindustrial, and poor countries. One answer might be this: Honest and literate customs inspectors are scarce in poor countries, but they are essential for implementing an ad valorem tariff where import valuation is so crucial. So, legislators impose specific duties to minimize the ‘‘theft’’ of state tariff revenues by dishonest and illiterate customs agents. Another answer might be this: Poor countries export primary products, concentrating on only a few, thus exposing themselves to price instability. Since export revenues and import expenditures are highly correlated, unstable export prices imply unstable export values and, finally, unstable tariff revenues. Specific duties tend to smooth out the impact of the export price instability on government finances. Controls: Policy Packages and Real Exchange Rate Trade-offs Few policies are decided in isolation. Indeed, there were other ways that governments could have improved the competitive position of import-competing industries, if such protection was their goal, and they explored many of these alternatives in the 1930s and in the ISI years that followed. Yet, they clearly understood these alternatives
Explaining World Tariffs, 1870–1938
215
even before World War I. One powerful alternative involved manipulating the real exchange rate. If governments chose to go on the gold standard or to peg to a core currency, they got more stable real exchange rates in return (and an attractive advertisement for foreign capital). However, since protection via real exchange rate manipulation was forgone, tariff rates would have to go up to reclaim that protection lost. Did countries exploit this trade-off both during the years of gold standard commitment before World War I and during the interwar years when everybody went off gold? The Political Economy of Tariff-Rate Setting: Empirical Analysis The potential explanations for tariff policies discussed in previous sections can be allocated to three main motives: strategic trade policy, revenue needs, and scarce factor tariff compensation. I take the infant industry development goal to be a mid- to late-twentieth-century motivation. These four central motives need not have been competing. Still, even though each may have played a role before World War II, we would like to know which played the biggest roles, and in which periods and places. Elsewhere, an econometric attack has been launched on the problem two ways (Blattman, Clemens, and Williamson 2002; Clemens and Williamson 2002; Coatsworth and Williamson 2004a): first, by treating the experience as comparative world economic history and thus exploring time series only (TS); and second, by exploring the cross-section variance across these thirty-five countries using time fixed effects (CS). The cross-section results are transformed to remove serial correlation (using the AR(1) Cochrane-Orcutt correction), and the time series are estimated using random effects (RE) after likewise correcting for serial correlation (with a Baltagi-Wu estimator). What I report here summarizes the main findings. Table 8.1 presents the time-series and cross-section results. Each of these contains five columns, necessitated by the fact that data coverage for inflation and the terms of trade is inferior to that of the other regressors. The right-hand side variables suggested by the previous section are the following (all but dummies in logs), allocated to the three central motives. Revenue Motive Lagged Export Share. This export/GDP ratio is a measure of export boom, where we expect booms in the previous year to diminish the
216
Table 8.1 Tariff Rate Determinants the World Around, 1870–1938 Dependent variable: In Own Tariff Includes AR(1) Baltagi-Wu (TS) or Cochrane-Orcutt (CS) serial correlation correction TS, country RE
Years Countries
1870–1938 All
1870–1938 All
1870–1938 All
1870–1938 All
1870–1938 All
1870–1938 All
1870–1938 All
1870–1938 All
1870–1938 All
1870–1938 All
0.0285 (1.36)
0.0832 (3.02)
0.0609 (2.30)
0.0463 (2.07)
0.0924 (3.32)
0.0397 (1.37)
0.0645 (1.67)
0.0601 (1.60)
0.0539 (1.80)
0.0753 (2.02)
0.2490 (9.06)
0.2507 (6.64)
0.2992 (8.45)
0.2246 (7.54)
0.2526 (6.67)
0.0440 (1.22)
0.0983 (1.82)
0.0338 (0.60)
0.0648 (1.76)
0.0953 (1.73)
0.0798 (2.22)
0.1219 (2.68)
0.1037 (2.55)
0.1371 (2.66)
Revenue Motive In Export Share
CS, year dummies
Strategic Tariff Motive In Partner Tariffs
Stolper-Samuelson Scarce Factor Compensation Motive In Terms of Trade Index In GDP per capita
0.1412 (2.40)
0.2227 (2.86)
0.1745 (2.28)
0.1810 (2.95)
0.2260 (2.90)
0.1025 (1.48)
0.1445 (1.44)
0.1228 (1.24)
0.1439 (2.00)
0.1435 (1.45)
In Schooling In Effective Distance
0.1640 (4.02) 0.0735 (4.86)
0.0560 (0.82) 0.1072 (4.95)
0.0573 (0.84) 0.1267 (5.97)
0.1719 (4.30) 0.0584 (3.76)
0.0416 (0.61) 0.1086 (5.02)
0.0672 (1.49) 0.0169 (0.74)
0.3046 (2.96) 0.0644 (1.53)
0.2993 (3.01) 0.0514 (1.28)
0.0548 (1.22) 0.0309 (1.29)
0.3053 (2.99) 0.0616 (1.48)
In Railway Mileage
0.0354 (3.38)
0.0639 (2.25)
0.0579 (1.98)
0.0347 (3.41)
0.0590 (2.08)
0.0055 (0.80)
0.0212 (0.93)
0.0190 (0.84)
0.0042 (0.56)
0.0219 (0.94)
In Urbanization
0.0478 (2.13)
0.0198 (0.30)
0.0013 (0.02)
0.0462 (2.10)
0.0235 (0.36)
0.0242 (0.99)
0.0890 (1.58)
0.0989 (1.66)
0.0211 (0.79)
0.0787 (1.41)
Historical Applications of Heckscher-Ohlin Theory
Specification
0.1224 (2.85)
0.0433 (0.84)
0.0545 (1.12)
0.1302 (3.00)
0.0504 (1.00)
Federal
0.0100 (0.35)
0.0524 (1.45)
0.0585 (1.55)
0.0071 (0.25)
0.0509 (1.35)
Colony
0.0033 (0.05)
0.1649 (0.83)
0.2797 (1.58)
0.0695 (1.50)
0.1515 (0.79)
In Population
0.1084 (2.50)
0.1716 (3.35)
Inflation Inflation Squared Constant N Groups Avg. obs/group R-squared overall DW original DW transformed
2.7797 (4.75) 2,138 35 61.1 0.224 0.222
5.8022 (7.80) 1,169 30 39 0.271 0.242
0.1441 (2.81)
0.1172 (2.58)
0.1721 (3.38)
0.0004 (1.45)
0.0005 (1.46)
0.0004 (0.90)
0.0003 (0.69)
0.0000 (2.45)
0.0000 (1.77)
0.0000 (0.44)
0.0000 (0.52)
5.4237 (7.45) 1,300 35 37.1 0.25 0.251
2.6333 (4.28) 1,951 35 55.7 0.251 0.227
5.1674 (6.68) 1,169 30 39 0.266 0.245
2,067
0.144 0.083 1.972
1,116
0.203 0.107 1.979
1,238
0.195 0.115 1.948
1,889
0.149 0.083 1.982
Explaining World Tariffs, 1870–1938
Controls
1,116
0.211 0.111 1.987
t-statistics are in parentheses below each coefficient estimate. War years (1914–1918) omitted. Schooling is measured as the number of people per 10,000 below the age of 15 who are enrolled in primary school. The terms of trade ¼ 100 in 1900. Source: Blattman, Clemens, and Williamson (2002), Table 3 (revised).
217
218
Historical Applications of Heckscher-Ohlin Theory
need for high tariff rates this year—if government revenues are a key motivation—thus yielding negative coefficients in the regression.11 Strategic Tariffs Motive Lagged Partner Tariffs. Strategic tariff policy suggests that countries should have imposed higher tariffs this year if they faced higher tariffs in their main markets abroad last year. Stolper-Samuelson Scarce Factor Compensation Motive Terms of Trade Index. In the periphery, this terms of trade variable measures the price of each jth country’s primary product exports ðPxjÞ relative to the price of manufactures ðPmÞ in world markets. In the core, the opposite is the case. If deindustrialization fears in the periphery were dominant, a positive coefficient should appear: price shocks in world markets that were good for the periphery’s export sectors were bad for import competing sectors inviting compensation for the injured parties. Thus, the sign on ln (Lagged Px=Pm) should tell us whether deindustrialization fears dominated in the periphery. In the European core and in land scarce Asia (like Japan), imports were dominated by foodstuffs and raw materials. Here, Px=Pm speaks to an ‘‘invasion of grain’’ fear, whether wheat or rice, inviting compensation for the injured parties in this case too; GDP per capita and Schooling, the latter the primary school enrollment rate. These variables are taken as proxies for skill endowments, with the expectation that the more abundant the skills, the more competitive the industrial sector, and the less the need for protection, thus yielding a negative coefficient in the regression; Effective Distance. The distance from each country to either the United States or the United Kingdom (depending on trade volume), adjusted by seaborne freight rates specific to that route. If protection was the goal, effective distance should have served as a substitute for tariffs, so the regression should yield a negative coefficient; Railway Mileage added in kilometers. Poor overland transport connections to interior markets serve as a protective device. Railroads reduce that protection, requiring higher tariffs to offset the effect. Thus, the regression should yield a positive coefficient;
Explaining World Tariffs, 1870–1938
219
Urbanization, taken as share of population in cities and towns greater than 20,000. This urbanization statistic is taken to be a StolperSamuelson proxy for the lobbying power of urban capitalists and artisans in the periphery, thus yielding a positive coefficient in the periphery regressions. Controls Inflation and inflation-squared. To the extent that countries used specific duties, inflation should have lowered tariff rates, thus yielding a negative coefficient. However, very rapid inflation might well have triggered a speedier legislative reaction with increases in specific duties, thus yielding a positive and offsetting coefficient on the squared term in the regression; Population. Large countries have bigger domestic markets in which it is easier for local firms to find a spatial niche. Alternatively, bigger populations imply higher density, which makes domestic tax collection easier and tariff revenues less necessary. In either case, the demand for protection should be lower in large countries, and the regression should produce a negative coefficient; Federal, a dummy variable; if a federal system ¼ 1, if centralized ¼ 0. Federal governments had a stronger need for customs duties ( joining members retained their tax authority), while centralized governments could better exploit internal revenue sources. Thus, the regression should report a positive coefficient; Colony, a dummy variable; if a ‘‘colony’’ ¼ 1; 0 otherwise.12 Comparative Tariff History Results Turning first to the time series in table 8.1, we see that all coefficients have the expected sign with the exception of schooling (at least some of the time). Revenue motivation is revealed since export booms were associated with lower tariffs. Backlash and compensation forces are revealed too, and in many ways. Decreases in overseas transportation costs were associated with an offsetting rise in tariff barriers, and increases in the length of the domestic rail network were associated with a symmetric rise in tariffs. As geographic barriers evaporated, import-competing industries were compensated by higher tariffs. Also,
220
Historical Applications of Heckscher-Ohlin Theory
an improvement in a country’s terms of trade in world markets generated a strong antiglobal reaction. For the periphery, this took the form of a deindustrialization reaction since an improvement in the relative price of their primary product export in world markets implied a fall in the relative price of imported manufactures, inviting a tariff-raising lobbying reaction by industrial interests at home. For the European core, this took the form of a grain invasion reaction, as a rise in the relative price of their manufacturing exports implied a fall in the relative price of their imported foodstuffs, inviting a tariff-raising lobbying reaction by landed interests at home. There is strong support for strategic tariff motives, since partner tariffs has a positive and significant coefficient throughout. The results for both schooling and urbanization depend on whether we control for inflation or not. Since including inflation reduces the sample size by almost half, however, we do not know if the different results for schooling and urbanization are due to the restricted sample or to the fact of controlling for inflation. In the full sample, an increase in urbanization was associated with an increase in tariffs, just as the Stolper-Samuelson theorem would predict, at least in the capital-scarce periphery. Tariff rates fell with increases in GDP per capita, a result consistent with modern surveys of global attitudes (O’Rourke and Sinnott 2003), but which I also find consistent with the Stolper-Samuelson theorem. Internal market size mattered in the predicted way: large countries had lower tariff rates. Finally, note that while inflation had the predicted effect throughout the period, it was not statistically significant. Thus, while inflation had its predicted effects during wartime (figure 8.1), it did not during peacetime. I believe this is due to the limited sample with inflation coverage underlying table 8.1.13 Judging by the estimated elasticities, increases in trading partner tariffs were by far the most powerful determinant of increases in own tariffs over the full seven decades, at least on the economic margin. Changes in GDP per capita, population, and schooling had elasticities next in size. The combined influence of geography—the sum of falling effective sealane distance and rising railway mileage—also had high elasticities, but they still were only half that of partner tariffs. The same is true of the terms of trade index. Much to my surprise, the lowest elasticity reported in table 8.1 is that attached to changes in the export share, suggesting that a revenue motive was not the dominant force after 1870. However, it was probably the dominant force before that date.
Explaining World Tariffs, 1870–1938
221
Having analyzed both statistical significance and marginal economic importance, what about historical significance? To see the difference, consider this example: Suppose that own tariffs were highly responsive to partner tariffs in the European periphery. Suppose also we observe that own tariffs rose in the European periphery. Was this antiglobal tariff rise due to changes in partner tariffs or some other force? We cannot answer this question without knowing how much partner tariffs changed. To pursue this example further, if partner tariffs barely changed, then we would have to look elsewhere for explanations of the historical rise in own tariffs despite the fact that for a given change in partner tariffs we see a large change in own tariffs. This case illustrates the difference between big marginal economic impact and big historical significance. Why were tariffs on the rise nearly everywhere in the decades before 1900? Elsewhere, it has been shown that growing GDP per capita and population size were serving to lower tariffs everywhere, but these were overwhelmed by tariff-raising forces (Blattman, Clemens, and Williamson 2002). The push for higher tariffs came mostly from two sources: first, domestic political economy forces associated with urbanization and schooling; and second, a protectionist reaction as a compensation to import-competing industries as openness was thrust upon them by advances in transportation technology (both on land and sea). Only in the European periphery do we observe partner tariffs making a major contribution to the antiglobal, tariff-raising dynamics during this period. Falling transportation costs certainly did contribute to rising tariff barriers in the European core, in the non-Latin European offshoots and in Asia. Yet, transport revolutions along the sealanes had little impact on tariffs in Latin America and the European periphery, simply because the fall in overseas freight rates were more modest there. In addition, there is strong evidence of deindustrialization fears in the periphery, joining deagricultural fears in the core. Overall, it appears to have been rising levels of railway penetration, schooling, urbanization (associated with changes in domestic politics) and improving terms of trade (at least up to the 1890s) that drove tariffs upward worldwide. For the period from the 1890s to World War I, those antiglobal domestic political economy and (dissipating) transportation forces pushing tariffs upwards were finally overwhelmed by surging proglobal forces: falling tariffs are associated with rising per capita incomes in Europe, their non-Latin offshoots and Latin America, carried in large
222
Historical Applications of Heckscher-Ohlin Theory
part by mass migrations and capital exports. Once again, the terms of trade effect was operative, but now in a proglobal way. As the longrun deterioration in the relative price of primary products (made famous by Raoul Prebisch) started after the 1890s, the relative rise in the price of imported foreign manufactures eased the competitive pressure on local industry in the periphery. During the interwar decades, the massive increases in tariffs were driven almost entirely by increases in partner tariffs, a force that seems to eclipse everything else. Cross-section Results Now consider the cross-section results in table 8.1. Here we control for two additional characteristics: colonial status—an indicator of autonomy over tariff policy, and federal status—an indicator of the decentralization of governance. Three variables appear to have a different impact in time series relative to cross-section: partner tariffs, schooling, and urbanization. The partner tariffs variable is not significant in cross-section and appears to be negative. How can this be consistent with a world in which, as we have seen, changes in a country’s own tariff is closely associated with changes in its trading partners’ tariffs? This cross-sectional pattern suggests that initial conditions were such that, before reacting to changes in their partners’ tariffs, countries began from a distribution in which high own tariffs just happened for other reasons to be associated with low partner tariffs and vice versa. This pattern would appear to fit Asia’s initial conditions at the dawn of the twentieth century: their own tariffs were forced to be low, either as colonies or as victims of gunboat diplomacy, while high tariffs prevailed in their American and European trading partners. The European periphery would appear to fit this characterization too: their backlash before World War I left them with high tariffs at a time when their trading partners in the European core had recently moved toward freer trade. This shows up in time series as a positive coefficient on trading partner tariffs, but the initial distribution of tariffs shows up in cross section as a negative coefficient. A similar argument can explain the predominantly negative (but insignificant) cross-sectional coefficient on urbanization. I have no explanation for the nonrobust coefficients on the schooling variable. What about historical significance? Why were tariffs so low before 1914 across Asia, the Middle East, and the European core? One reason was the large internal markets in these labor-rich and land-scarce
Explaining World Tariffs, 1870–1938
223
economies. Another was the industrial competitiveness of the European core as captured by GDP per capita. Why were tariffs so high in both Latin and non-Latin European offshots? It appears that smaller domestic markets in the Latin and non-Latin European offshoots made it harder for firms to survive in a niche without walls to protect them, and, of course, they were less competitive. While the revenue motive is certainly present, and while the signs and magnitudes on the export share coefficient are the same in cross-section and time series, the influence is less powerful in the cross-section, a result I find surprising. Where Do We Go from Here? While surveying the political economy of trade policy, Dani Rodrik concluded that the ‘‘links between the empirical and theoretical work have never been too strong’’ (Rodrik 1995:1480). It appears that the boom in endogenous tariff theory over the past two decades has far outstripped the evidence brought to bear on it. My hope is that comparative tariff history like that offered in this essay will help redress the balance, and, in so doing, provoke new thinking on the political economy of tariffs. This essay relies on a data base documenting average tariffs between 1865 and 1938 for thirty-five countries. While tariff policy for industrial Europe and the United States has been studied extensively, the rest of the world has not, and of our sample of thirty-five, the majority are from the periphery: ten are from the European periphery; another ten are from Asia and the Middle East; and the remaining eight are from Latin America. The advantage of this large panel database is obvious since it documents an enormous range of tariff policy experience, by period and by country. What accounts for this immense variety in both cross section and time series? What were the underlying fundamentals driving tariff policy the world around? I think these questions should be at the top of the international economist’s agenda. After all, even if we see high and rising tariffs out there in history, we need to know why they were high and rising if this history is to be used to understand the future of globalization in the present century. We have learned a fair amount in this essay: deindustrialization fears were a major determinant of tariff policy in the periphery before World War I, joining grain-invasion fears in the European core; revenue needs were an important determinant of tariff rates in the periphery, and especially for young republics;
224
Historical Applications of Heckscher-Ohlin Theory
geography mattered, so that where and when the natural protection of distance and topography was conquered by transport technology, tariffs rose to compensate the import competing industries; and, finally, there was strategic tariff policy behavior at work everywhere after World War I, and much earlier in most young republics forming customs unions. There is much more to be done to uncover the fundamentals driving world tariff policy in the century before World War II. And while economic historians are doing it, one can only hope that economists will do the same for recent experience. Eli Heckscher would be pleased. Acknowledgments This project has received superb research assistance from David Clingingsmith, Martin Kanz, and Istva´n Zo¨llei. I have also benefitted by comments from Dick Baldwin, Gerry della Paolera, Graciela Marquez, Richard Cooper, Toni Estevadeordal, Ron Findlay, Jeff Frieden, Steve Haber, Elhanan Helpman, Doug Irwin, Jonas Ljungberg, Matts Lundahl, Ed Leamer, Peter Lindert, Stephen Meardon, Jose´ Manuel Ocampo, Kevin O’Rourke, Jonathan Pincus, Leandro Prados, Dani Rodrik, Dick Salvucci, Ken Sokoloff, Alan Taylor, Daniel Waldenstro¨m, and Nick Wills-Johnson; participants at the Conference on the Political Economy of Globalization (Dublin, August 29–31, 2002), the Economic History Association Meetings (St. Louis, October 11–13, 2002), the FTAA and Beyond Conference (Punta del Este, December 15–16, 2002), the RIN Conference (Punta del Este, December 17–19, 2002); and seminars at Adelaide, ANU, Australian Productivity Commission, Australian Treasury, Curtin, Harvard, UCLA, and UC-Davis. Most important, this essay draws extensively on a tariff project which has involved previous papers with four collaborators: Luis Be´rtola, Chris Blattman, John Coatsworth, and Michael Clemens. I am grateful to all of them, but errors remaining belong to me. I also acknowledge generous financial support from the National Science Foundation SES0001362. Notes 1. The average tariff rate is measured here as the share of customs revenues (import duties only) in total import values. It is part of the Williamson Tariff Project data base, used with collaborators in a series of papers (Be´rtola and Williamson forthcoming; Clemens and Williamson 2002, 2004; Coatsworth and Williamson 2004a; Blattman,
Explaining World Tariffs, 1870–1938
225
Clemens, and Williamson 2002). Toni Estevadeordal of the Inter-American Development Bank and I are now constructing a database documenting tariff structure in Latin America since the 1820s. 2. I have also calculated (but do not report here) weighted tariff averages for the regional clubs in figure 8.2, where weights are the country’s total export share in regional exports or its GDP share. However, I prefer to treat countries as independent policy units regardless of size. 3. Of course, embargoes and soaring transport costs served to produce the same, or even bigger, protective effects. 4. The United States has always presented a problem to historians and economists alike. The canonical frontier economy with scarce labor and abundant resources, by 1900 it was also the world’s industrial leader (Wright 1990) and a central market for the exports from the rest of the world, especially Latin America. So, while the United States was certainly a rich European offshoot, I allocate it to the industrial core. 5. In 1831, only 8.6 percent of the males in the United Kingdom had the right to vote, and even in 1866, after the First Reform Act in 1832, the figure was still only 17.8 percent. (See Lindert 1998: table 4.) These were, of course, the wealthy at the top of the distribution. 6. As late as 1940, the share of the population voting in Latin America was never higher than 19.7 percent (Uruguay), while the lowest figures were for Ecuador, Bolivia, Brazil, and Chile (3.3, 4.1, 5.7, and 6.5 percent, respectively). Engerman, Haber, and Sokoloff (2000:226, table 2). 7. Industrial manufactures have been a rapidly rising share of Third World output and exports. For example, for all ‘‘developing’’ countries, manufactures rose from only 17.4 percent of commodity exports in 1970 to 64.3 percent by 1994. Enough of the Third World is now labor-abundant and natural resource–scarce so that the growth of trade has helped it industrialize. The classic image of Third World specialization in primary products is obsolescing. See Lindert and Williamson (2003, note 22). 8. Caution suggests using the phrase ‘‘was associated with’’ rather than ‘‘fostered.’’ I press on without caution, but subject to this understanding. 9. Any evidence favoring the Stolper-Samuelson hypothesis would, of course, also support the backlash view. 10. Except, of course, that they could keep the price of imported raw material intermediates low by giving such imports tariff concessions. 11. In a related paper on Latin America (Coatsworth and Williamson 2004a), capital inflows from Britain were added to the analysis for the years 1870–1913. This variable measured annual British capital exports to potential borrowing countries. Countries favored by British lending were shown to have had less need for tariff revenues and thus had lower tariffs. The variable does not appear here since our source does not report the period 1914–1938. Similarly, I do not report the gold standard effect here, although we now have the data to report the answer: being on the gold standard was associated with higher tariff rates, as predicted. 12. New analysis not reported here replaces the ‘‘colony’’ dummy with a ‘‘tariff autonomy’’ dummy—to include Asian countries with unequal treaties—and the results are even stronger than those reported in table 1.
226
Historical Applications of Heckscher-Ohlin Theory
13. Ongoing work is repairing this data limitation, and the specific-duty effect is confirmed in the larger sample. In addition, recall that the analysis in table 8.1 excludes World War I, and we know these years to have been ones during which the specificduty effect was powerful.
References Bagwell, K., and R. W. Staiger. 2002. The Economics of the World Trading System. Cambridge, MA: MIT Press. Bairoch, P. 1989. ‘‘European Trade Policy, 1815–1914.’’ In P. Mathias and S. Pollard, eds., The Cambridge Economic History of Europe, 3. Cambridge: Cambridge University Press. Be´rtola, L., and J. G. Williamson. forthcoming. ‘‘Globalization in Latin America Before 1940.’’ In V. Bulmer-Thomas, J. Coatsworth, and R. Corte´s Conde, eds., Cambridge Economic History of Latin America. Cambridge: Cambridge University Press. Bhagwati, J. N. 2000. Free Trade Today. Princeton, NJ: Princeton University Press. Blattman, C., M. A. Clemens, and J. G. Williamson. 2002. Who Protected and Why? Tariffs the World Around 1870–1938. Paper presented to the Conference on the Political Economy of Globalization, Trinity College, Dublin, August 29–31. Bulmer-Thomas, V. 1994. The Economic History of Latin America Since Independence. Cambridge: Cambridge University Press. Clemens, M. A., and J. G. Williamson. 2002. ‘‘Closed Jaguar, Open Dragon: Comparing Tariffs in Latin America and Asia before World War II.’’ NBER Working Paper 9401, National Bureau of Economic Research, Cambridge, MA. ———. 2004. ‘‘Why Did the Tariff-Growth Correlation Reverse After 1950?’’ Journal of Economic Growth 9: 5–46. Coatsworth, J. H., and J. G. Williamson. 2004a. ‘‘The Roots of Latin American Protectionism: Looking Before the Great Depression.’’ In A. Estevadeordal, D. Rodrik, A. Taylor, and A. Velasco, eds., FTAA and Beyond: Prospects for Integration in the Americas 37–73. Cambridge, MA: Harvard University Press. ———. 2004b. ‘‘Always Protectionist? Latin American Tariffs from Independence to Great Depression.’’ Journal of Latin American Studies 36:205–232. Conybeare, J. 1983. ‘‘Tariff Protection in Developed and Developing Countries: A CrossSectional and Longitudinal Analysis.’’ International Organization 37, no. 3 (Summer): 441–467. Corbo, V. 1992. ‘‘Development Strategies and Policies in Latin America: A Historical Perspective.’’ International Center for Economic Growth, Occasional Paper No. 22 (April): 16–48. Crucini, M. J. 1994. ‘‘Sources of Variation in Real Tariff Rates: The United States 1900– 1940.’’ American Economic Review 84 ( June): 732–743. Diaz-Alejandro, C. 1984. ‘‘Latin America in the 1930s.’’ In R. Thorp, ed., Latin America in the 1930s, 17–49. New York: Macmillan. Dixit, A. 1987. ‘‘Strategic Aspects of Trade Policy.’’ In T. F. Bewley, ed., Advances in Economic Theory: Fifth World Congress, 329–362. New York: Cambridge University Press.
Explaining World Tariffs, 1870–1938
227
Engerman, S., S. Haber, and K. Sokoloff. 2000. ‘‘Institutions, Factor Endowments, and Paths of Development in the New World.’’ Journal of Economic Perspectives (Summer 2000): 217–232. Grossman, G., and E. Helpman. 1994. ‘‘Protection for Sale.’’ American Economic Review 84, no. 4 (September): 833–850. Heckscher, E. F. 1991. ‘‘The Effect of Foreign Trade on the Distribution of Income.’’ In H. Flam and M. J. Flanders, eds., Heckscher-Ohlin Trade Theory, 39–69. Cambridge, MA: MIT Press. Irwin, D. A. 1998a. ‘‘Higher Tariffs, Lower Revenues? Analyzing the Fiscal Aspects of the Great Tariff Debate of 1888.’’ Journal of Economic History 58 (March): 59–72. ———. 1998b. ‘‘Changes in U.S. Tariffs: The Role of Import Prices and Commercial Policies?’’ American Economic Review 88 (September): 1015–1026. Kindleberger, C. P. 1951. ‘‘Group Behavior and International Trade.’’ Journal of Political Economy 59 (February): 30–46. Krugman, P. 1996. Pop Internationalism. Cambridge, MA: MIT Press. Lindert, P. H. 1994. ‘‘The Rise in Social Spending, 1880–1930.’’ Explorations in Economic History 31 ( January): 1–36. ———. 1998. ‘‘Poor Relief Before the Welfare State: Britain Versus the Continent 1780– 1880.’’ European Review of Economic History 2 (1998): 101–140. Lindert, P. H., and J. G. Williamson. 2003. ‘‘Does Globalization Make the World More Unequal?’’ In M. Bordo, A. M. Taylor, and J. G. Williamson, eds., Globalization in Historical Perspective, 227–271. Chicago: University of Chicago Press. Magee, S. P., W. A. Brock, and L. Young. 1989. Black Hole Tariffs and Endogenous Policy Theory. Cambridge: Cambridge University Press. Ma´rquez, G. 2002. ‘‘The Political Economy of Mexican Protectionism, 1868–1911.’’ Ph.D. thesis, Harvard University. Marvel, H. P., and E. J. Ray. 1983. ‘‘The Kennedy Round: Evidence on the Regulation of International Trade in the USA.’’ American Economic Review 73 (March): 190–197. ———. 1987. ‘‘Intraindustry Trade: Sources and Effects on Protection.’’ Journal of Political Economy 95 (December): 1278–1291. McGillivray, F., I. McLean, R. Pahre, and C. Schonhardt-Bailey. 2001. ‘‘Tariffs and Modern Political Institutions: An Introduction.’’ In F. McGillivray et al., eds., International Trade and Political Institutions: Instituting Trade in the Long Nineteenth Century, 1–28. Cheltenham, UK: Edward Elgar. Mill, J. S. 1909. Principles of Political Economy. London: Longmans. Mohammed, S. S., and J. G. Williamson. 2004. ‘‘Freight Rates and Productivity Gains in British Tramp Shipping 1869–1950.’’ Explorations in Economic History 41:172–203. Obstfeld, M., and A. M. Taylor. 2003. ‘‘Globalization and Capital Markets.’’ In M. Bordo, A. M. Taylor, and J. G. Williamson, eds., Globalization in Historical Perspective, 121–187. Chicago: University of Chicago Press.
228
Historical Applications of Heckscher-Ohlin Theory
O’Rourke, K. H. 1997. ‘‘The European Grain Invasion, 1870–1913.’’ Journal of Economic History 57 (December): 775–801. O’Rourke, K. H., and R. Sinnott. 2003. ‘‘Migration Flows: Political Economy of Migration and the Empirical Challenges.’’ Unpublished paper (March). O’Rourke, K. H., and J. G. Williamson. 1999. Globalization and History. Cambridge: Cambridge University Press. Pincus, J. 1977. Pressure Groups and Politics in Antebellum Tariffs. New York: Columbia University Press. Rodrik, D. 1995. ‘‘Political Economy of Trade Policy.’’ In G. M. Grossman and K. Rogoff, eds., Handbook of International Economics 3:1457–1494. Amsterdam: Elsevier. Rogowski, R. 1989. Commerce and Coalitions: How Trade Effects Domestic Political Arrangements. Princeton, NJ: Princeton University Press. Smith, A. 1776 (1976). An Inquiry into the Nature and Causes of the Wealth of Nations. Oxford: Clarendon Press. Stolper, W., and P. Samuelson. 1941. ‘‘Protection and Real Wages.’’ Review of Economic Studies 9:58–73. Williamson, J. G. 1998. ‘‘Globalization, Labor Markets and Policy Backlash in the Past.’’ Journal of Economic Perspectives 12 (Fall): 51–72. ———. 2002. ‘‘Two Centuries of Globalization: Backlash and Bribes for the Losers.’’ WIDER Annual Lecture, Copenhagen, September 5. Wright, G. 1990. ‘‘The Origins of American Industrial Success, 1879–1940.’’ American Economic Review 80 (September): 651–668.
Part IV
Mercantilism
9
Eli Heckscher and His Mercantilism Today Lars Magnusson
There are two possible approaches to assessing mercantilism as a phenomenon, as well as to evaluate Eli Heckscher’s contribution to our understanding of it. The first of these is that of mainstream economics: that is, to examine the doctrines and policies that constituted mercantilism, as well as the work of Heckscher and other interpreters of mercantilism, from the perspective of modern economic theory. This approach generates questions such as the following: Did the mercantilists, as Keynes argued, achieve economic insights that have been lost? What understanding can be gained by applying a public choice inspired (rent-seeking) analysis to seventeenth- and eighteenth-century mercantilist policies? Is it fruitful to utilize modern concepts, such as transaction cost and asymmetrical information, in our quest to understand the historical phenomenon of mercantilism? How, if at all, can Heckscher’s, or Keynes’s, reinterpretation of mercantilist ideas be reconciled with modern concepts and theories? The second possible strategy is historical in nature. It emphasizes the necessity of placing the phenomenon, as well as its interpretation, in the proper historical context. The question of what constituted mercantilism and how it has been interpreted over time can be understood only in terms of its meaning during various historical epochs, including how it was interpreted by those who lived during the time when it was conceived and advocated. Thus, for example, the practices of state dirigisme are incomprehensible without a knowledge of the particular institutional setting of the time, including undeveloped markets and sticky factor prices. By the same token, any interpretations of ideas (or policies) also have their own history and can only be understood within that context. Both approaches are clearly useful, although for different purposes, and they should not be treated as being totally contradictory. That,
232
Mercantilism
however, has not prevented modern (public choice) economists, such as Robert B. Ekelund and Robert D. Tollison, from scorning the use of the historical method—among whose practitioners they include Heckscher—as a ‘‘red-herring of historiography’’ (Coleman 1969:117), which ‘‘has yielded no cogent or consistent result if its aim has been a definition of mercantilism or an understanding of the period’s events and institutional change’’ (Ekelund and Tollison 1997). According to them, historical interpretations invariably result in a ‘‘mess.’’ Of course, it can be countered that the economist’s method of reading history backward also has serious disadvantages. It frequently leads to anachronistic conclusions and judgments concerning historical pamphleteers and authors that miss the crucial point. That is, what it was that they actually were trying to say within their own historical context. First of all, there is a tendency to headline thinkers and writers who were not well known as economists and who played no role in either the scholarly or the public debate of their time. This might not be a problem for understanding the development of pure analysis. It is a major difficulty, however, in trying to understand the broader history of economics. Second, there is a danger that the purely economic approach will lead to subjectivism. Thus an author’s importance—no matter how historically obscure he may be—is measured by the extent he has contributed to, or been in tune with, the most recent version of economic theory (see, for example, Rothbard 1995). Third, there is a danger that concepts, and even theories, might be reinterpreted in a fashion that obscures their original meaning. This makes it more difficult for us to understand what ideas the mercantilist writers wished to convey and why they presented them as they did. Thus, for example, twisting the arguments of earlier economists to make them seem to be precursors of modern concepts and theories is of doubtful value. The results of such an exercise are usually predetermined. We find what we wish to find. Moreover, doing so makes it harder to learn what there actually is to learn from these authors, namely that there are various ways of looking at economic problems and that economics as a subject is historically embedded. Surely the history of economic thought includes more than the logical perfection of what Arthur Lovejoy once called ‘‘unit ideas’’? It also deals with how various individuals, be they rulers, merchants, or intellectuals, have tried to understand, and come to grips with, a complex and continually changing economic order, as well as with how these actors have distinguished what is good from what is bad in their world.
Eli Heckscher and His Mercantilism Today
233
Eli Heckscher and Mercantilism The first Swedish edition of Eli Heckscher’s Mercantilism appeared in 1932 (Magnusson 1994b). It was translated into German the following year, and the first English edition was published in 1935. Virtually overnight, it made Heckscher well known to a wide international audience. However canonical this work seems now, when it was first published it received a mixed response from both economists and economic historians. The book’s reviewers acknowledged the massive scholarly effort that had gone into writing it, as well as Hecksher’s great learning and skill. Prominent economic and social historians, such as Marc Bloch (1934) and Herbert Heaton (1937), however, were highly critical of Heckscher’s conceptualization of a ‘‘system of mercantilism.’’ Bloch thought it had an unhistorical aura about it, and Heaton emphasized Heckcher’s difficulty in establishing a synthesis among ‘‘the situation, the ideas and the actions’’ of mercantilism (Magnusson 1994a:32–36). Moreover, he found it hard to accept the argument that all the regulative policies practiced by governments starting in the Middle Ages had been based on a common and coherent set of intentions and goals. Finally, Heckscher’s notion of a ‘‘fear of goods’’ as the principal explanation for the rise of mercantilism and as a reflection of the transition to monetized economies in Western Europe was judged to be both overly general and unrealistic. It is certainly true that Heckscher interpreted mercantilism very broadly. In fact he treated it as a system of economic, regulatory, administrative, and political thinking with roots going back to the policies of the medieval cities. Thus, in essence, mercantilism was ‘‘a phase in the history of economic policy’’ (Heckscher 1994, 2:2). It was also, however, a systematic doctrine: a ‘‘system of protection and money.’’ As such, he treated it as a species of timeless commonsensical, popular economic thinking. According to Heckscher, it was not limited to any specific historical period. But even this was not all. Mercantilism was also a particular conception of man and society: almost a (extremely materialistic) worldview. Moreover, it fitted neatly into a stages of history model of the type advocated by the historical school—an approach of which Heckscher himself was critical (Heckscher 1942). Heckscher’s historicist tendencies are also displayed in his suggestive notion of a ‘‘fear of goods’’: that the hunger for money, taken to the level of a fetish, together with the corresponding ‘‘fear of goods,’’ was a refection of the transition from a barter to a money economy.
234
Mercantilism
One important reason for the cool reception given Heckscher’s book was no doubt its complex structure. Thus, for example, it is often difficult to grasp how its various parts relate to each other. To a large extent, this problem is a product of Heckscher’s own ambivalence concerning the nature of mercantilism. Indeed, this ambivalence is the principal explanation for his eclectic, unsynthesized treatment of the subject. He deals with various features of mercantilism, such as regulatory policies, economic doctrines, and general conceptions of society, separately and without clarifying how they relate to each other. Heckscher’s general argument, however, can be reconstructed as follows: He begins by emphasizing the systemlike character of mercantilism as a set of specific economic policies. He then asserts, as a matter of principle, that ‘‘everybody has certain ideas, whether he is conscious of them or not, as a basis for his actions, and mercantilists were plentifully provided with economic theories on how the economic system was created and how it could be influenced in the manner desired’’ (Heckscher 1994, 1:27). Furthermore, he suggests, mercantilism can only be understood by separating its ends from its means. The ultimate goal of mercantilist policies was to maximize the external power of the state. This explicitly contradicted Adam Smith and the liberal economists, who placed the wealth of the individual ahead of that of the nation state. This goal, however, was not mercantilism’s most distinctive characteristic. Rather, Heckscher emphasized that the hallmark of the system was the set of means intended to achieve this general goal. These economic means designed to bolster the political strength of the state were inherent to mercantilism as a protectionist and monetary system. The ambiguity as to whether this ‘‘system’’ should be classified as economic policy or economic thinking—or both—is not initially resolved and haunts the reader throughout the two massive volumes. Regardless of whether it was policy or doctrine, however, Heckscher stresses that mercantilism must not be seen as a rational reflection of how economic systems functioned during the early modern period. Thus, for example, he points out that ‘‘the description of the economic policy pursued in a particular period should never be regarded as a sufficient explanation of the economic circumstances of the time’’ (Heckscher 1994, 1:20). In a chapter added to the second edition, he is even more explicit. Now, however, he has reversed the argument to deal with the economic thinking of the period. His principal assertion is well known: ‘‘There are no grounds whatsoever for supposing that
Eli Heckscher and His Mercantilism Today
235
the mercantilist writers constructed their system—with its frequent and marked theoretical orientation—out of any knowledge of reality however derived’’ (Heckscher 1955, 2:347). On the basis of this conceptualization, Heckscher informs the reader that he will deal with five different aspects of mercantilism. First, there is mercantilism as a policy of national unification and integration. Second, there is the desire for national power. The third and fourth aspects are protectionism and the monetary system. Here Heckscher takes his point of departure from Adam Smith. Finally, there is the concept of society inherent in mercantilism. As Heckscher observes, this last aspect has frequently been neglected. On the basis of this foundation, Heckscher’s overall aim is to provide a synthesis of these five aspects and thus to provide a general interpretation of mercantilism as a systematic phenomenon. The Economists’ Critique of Heckscher It is well known that the concept of a ‘‘mercantile system’’ was invented by the French physiocrats and subsequently made famous by Adam Smith. During the late nineteenth century, both in Germany and Britain, there was a reaction against the portrayal of mercantilism as being little more than a set of logical errors based on the false belief that money was identical with wealth. Instead, historical economists, such as Schmoller in Germany and Cunningham in Britain, argued that mercantilism had a rational basis. According to them, mercantilism could better be understood as a broad process of state, or nation, building. Thus, not only did they reinterpret mercantilism as a broad concept applicable to certain types of policy making and economic management of the state during the early modern period, they also brought into serious question Smith’s assertion that mercantilism must be regarded as a doctrine pursued for the benefit of special, private interest groups. Instead, these revisionists argued that, at its core, mercantilism represented the interests of the national state. First and foremost, it reflected the ‘‘economic interests of whole states . . . [which] found a rallying-point in certain generally accepted postulates’’ (Schmoller 1896:59). In ‘‘national policy’’ terms, the particular views and ideas of individual mercantilist thinkers and writers were of little interest. Thus, for example, ‘‘the whole idea and doctrine of the Balance of Trade . . . was only the secondary consequence of economic processes’’ (Schmoller 1896:61). Instead, the true basis of mercantilism was
236
Mercantilism
the workings of particular institutions in given countries and the goals of statesmen and rulers. Even though Heckscher was not always prepared to admit it, there is no doubt that his interpretation of mercantilism was greatly influenced by the historical economists. Thus, in accord with Schmoller and the historical school, he expanded the term ‘‘mercantilism’’ to be the kernel of a system of economic thought and a concept of society, not just a system of economic policy stretching back to the Middle Ages. Following in the footsteps of, among others, Schmoller and Cunningham, Heckscher made the state the principal agent of, and the chief propelling force behind, mercantilism. Although he conceded that opulence, as well as power, was indeed a goal of mercantilist policies, there is certainly some truth in Heaton’s judgment that Heckscher ‘‘insists that mercantilism put power above opulence, in contrast with laissez-faire, which made the creation of wealth its lodestar, with small regard to the effect on power of the state’’ (Heaton 1937:379). In a famous two-part essay published in 1930, Jacob Viner took the side of Adam Smith against the historical economists. Instead of trying to understand the mercantilists on their own terms, Viner maintained that one should start with ‘‘modern monetary and trade theory’’ and on this basis provide ‘‘an inventory of English ideas with respect to trade prevalent before Adam Smith, classified and examined in the light of modern . . . theory’’ (Viner 1930:250). He believed that such a critical evaluation of the evolution of doctrines would yield a greater understanding of them than would the historicists’ approach, which was based on the belief that these ideas reflected the economic reality of the premodern world. Viner’s argument, incidentally, is almost exactly echoed by Ekelund and Tollison in their current critique of the historical method (1997). According to Viner, the mercantilists erred exactly as Adam Smith had claimed: they confused real wealth with money. There is no other way, he stressed, to explain the favorable balance of trade doctrine or their belief that foreign trade was the only possible path to national enrichment. Consequently, he maintains, the mercantilists ‘‘believed, momentarily at least, that all goods other than money were worthless’’ (Viner 1930:265). By relying on quotes taken out of their proper textual context, Viner is thus able, even more fervently than Adam Smith, to argue that at heart the mercantilists were out and out bullionists. At the same time, however, he was prepared to admit that the total identification of money with wealth was limited to some extreme mercanti-
Eli Heckscher and His Mercantilism Today
237
lists. The identification of gold and silver as the only ‘‘real’’ form of wealth by their more moderate compatriots must be understood from an institutional and historical perspective. It is thus ironic that in his discussion regarding why money was considered so pivotal during the sixteenth and seventeenth centuries, Viner accepted an explanation which stressed the ‘‘material’’ (or institutional) basis of this fallacy. In this context, he maintained that ‘‘much more important in the writings of the abler mercantilist than the absolute identification of wealth with gold and silver was the attribution to the precious metals of functions of such supreme importance to the nation’s welfare as to make it proper to attach to them a value superior to that of other commodities’’ (Viner 1930:270). In addition, Viner challenged another assumption made by the historical economists, namely that the mercantilist writers were disinterested and were arguing for the general good when they defended government policies and legislation. According to Viner, the opposite was true. They were first and foremost advocates of ‘‘special interests’’ (using modern terminology, they would be labeled rent seekers). Each interest group continually lobbied for legislative action in their own favor. Viner’s pair of articles appeared before Heckscher’s work had been published and translated. Reading Mercantilism, however, only reinforced Viner’s critical attitude toward historical economics. Thus, in his review of the book, as well as in an article published a decade later, he severely criticized the notion that power had been an end in itself for the mercantilists (Viner 1935, 1948). A more charitable reading of Heckscher, however, suggests that he saw power as only one of several mercantilist goals. After all, his intention was to present a general synthesis in which the struggle for power was just one aspect. Moreover, in principle Heckscher was always reluctant to advocate monocausal explanations. In his response to Viner, he was therefore prepared to grant that both power and opulence were central themes in mercantilist economic policy during its heyday. Furthermore, he suggested, underlying these two linked goals lurked a very peculiar social philosophy: the worldview of mercantilism. The Response from the Economic Historians The 1950s and 1960s were characterized by the rapid rise of economic history as an academic enterprise, especially in the Anglo-Saxon world. Undoubtedly, rapid industrial transformation, a high growth rate, and
238
Mercantilism
controversy over social issues lay behind this increase in its popularity. In Britain, however, one major theme of economic historical scholarship concerned mercantilism. This discussion focused on two issues in particular. The first of these was the usefulness of Heckscher’s broad and encompassing definition of the concept. The consensus on this question was generally negative. The second issue concerned the impact of new research on international trade relations on our understanding of the phenomenon of mercantilism. Here, a lively discussion intended to establish a sound economic and historical foundation for the superstructure of mercantilist thinking blossomed. It is only possible to speculate as to why mercantilism became such a popular topic during these decades. Very likely, the free trade versus protectionism controversy that had been ongoing in Britain since the nineteenth century was one factor. The ‘‘mercantilist fallacy,’’ that a positive balance of trade was the road to national wealth, was perceived by many free traders—including Viner in America and Heckscher in Sweden—to be a useful weapon against protectionist theories. If mercantilism and protectionism could be shown to be built on a logical fallacy, it would be even easier to demonstrate the superiority of free trade (Magnusson 2004). With regard to Heckscher’s ‘‘totalitarian’’ ambitions, as early as 1939 A. V. Judges had vigorously rejected the notion of ‘‘a mercantile state.’’ Formally, his campaign was directed against the Historismus of ‘‘German scholarship,’’ including British fellow travellers such as William Cunningham and W. J. Ashley. Although he was only briefly mentioned, however, Judge’s criticisms could just as well have been directed at Heckscher. Judges raised the question of whether or not it was appropriate to classify mercantilism as a coherent ‘‘system.’’ Of course, he was certain of his own answer. Mercantilism ‘‘never had a creed; nor was there a priesthood dedicated to its service,’’ he wrote. Moreover, he argued, it did not present a coherent doctrine, or ‘‘at least a handful of settled principles.’’ Thus, mercantilism was a straw man constructed ‘‘in the eighteenth century by men who found security for their own faith in a system of natural law’’ ( Judges [1939] 1969:35f). Some two decades later, this standpoint was further elaborated by a leading British economic historian, D. C. Coleman, who directly confronted Heckscher’s ‘‘synthesizing treatment’’ of mercantilism. In Heckscher’s Hegelian-inspired hands, mercantilism had become a real entity, manifesting itself through the centuries in a variety of guises. Coleman’s conclusion is widely accepted: ‘‘what was this mercantilism.
Eli Heckscher and His Mercantilism Today
239
Did it exist? As a description of a trend of economic thought the term may well be useful. . . . As a label for economic policy, it is not simply misleading but actively confusing, a red-herring of historiography. It serves to give a false unity to disparate events, to conceal the close-up reality of particular times and particular circumstances, to blot out the vital intermixture of ideas and preconceptions, of interests and influences, political and economic, and of the personalities of men which is the historian’s job to examine’’ (Coleman 1969:116; see also Earle 1950). As already noted, in subsequent articles Coleman has extended this argument to include mercantilism as a trend in economic thought. In 1980 he conceded that the term mercantilism might have a certain heuristic value. Indeed, it was an example of such ‘‘non-existent entities that had to be invented in order to prevent the study of history of falling into the abyss of antiquarianism.’’ As a description of a real specific stream of economic thought or economic policy, however, he maintained that the term was illegitimate and misleading. A second important theme in the discussion of mercantilism since the 1950s has dealt with the relationship among economic ideas, events, and policies during the ‘‘age of mercantilism’’ (roughly the seventeenth and eighteenth centuries). Although Heckscher, as argued above, believed in a clear connection between thought and policy, he emphatically rejected the notion of a causative relationship between thought and events. In this regard, he was severely criticized by, among others, Coleman. Conversely, Coleman emphasized the materialistic foundation of economic ideas. Thus, the basis for ‘‘mercantilist’’ ideas must be sought in the historical reality of the seventeenth and eighteenth centuries (Coleman 1969:111). Consequently, starting in the 1950s, Coleman’s economic history colleagues set out to uncover the historical basis and origins of the theories regarding trade and growth that flourished before Adam Smith. The question they posed concerned the real historical foundation on which the writings of the mercantilists rested. Much historical research was devoted to this quest. Various scholars came up with alternative answers and interpretations. Generally speaking, these contributions have produced a much fuller understanding of the prevailing economic structures and conditions which are thought to have been the foundation of the mercantilist ideas and texts. As an example, Charles Wilson has argued that a common weakness in attacks against the mercantilists, from Smith onward, has been their
240
Mercantilism
denial of ‘‘the possibility that the obsession with bullion might have rational historical roots’’ (Wilson 1969:64). Wilson, however, maintains that a central theme in early British mercantilist writings (Mun, Misselden, and the like) was concern over particular trade balances, especially in the Anglo-Baltic trade. As Wilson observes, much trade during this period was bilateral in nature. The importation of grain, timber, iron, and copper from the Baltic countries was crucial for Britain. It was an ongoing problem, however, that this trade required the exportation of metallic money from Britain. An important reason for this situation was the Dutch domination of exports to the Baltic area (Wilson 1957). Consequently, this unavoidable outflow of specie had to be offset by a positive ‘‘overplus’’ in the British trade with other countries and regions. Wilson believed that it was this rather special circumstance that explains the mercantilists’ concern with maintaining a positive balance of trade. As trade became ever more multilateral during the seventeenth century, this doctrine became increasingly outdated and was ultimately discarded (Wilson 1949). In his response to Wilson, Heckscher denied that bilateral trade was common during the seventeenth century. According to Heckscher, the existence of exchange bills argued strongly against Wilson’s explanation (Heckscher 1950). In a rejoinder, Wilson reasserted his view that ‘‘in a number of branches of international trade, precious metals played a unique role which gives an element of rationality to mercantilist thought’’ (Wilson 1951–52:242; see also Price 1961). Moreover, the ‘‘hard currency’’ position was ‘‘rooted in the views of individual merchants about the requirements of their business.’’ Therefore, he concluded, ‘‘Trading capital in money was regarded as an indispensable link in the exchange of goods’’ (Wilson 1951–52:54). During the early 1950’s, another line of research, also openly intended to endow the mercantilist writers with historical rationality, became the focus of intense debate. In two articles published in 1954 and 1955, J. D. Gould argued that the great British commercial and industrial depression of the 1620s played a major formative role in the emergence of mercantilist doctrines. According to Gould, most contemporary observers agreed that the crisis primarily was caused by ‘‘some shortcoming of the monetary system and the machinery of foreign exchange’’ (Gould 1954:82). From this it followed that: ‘‘To one who studied the trade depression of the 1620s in detail . . . it is clear that a very substantial part of Englands’s Treasure [Mun’s most famous book] represents simply the fruits of reflection on the events and dis-
Eli Heckscher and His Mercantilism Today
241
cussion of those years’’ (Gould 1955:123). In referring to the monetary situation, Gould unquestionably drew on De Roover’s path-breaking study of Thomas Gresham and international exchange relations during the sixteenth century (De Roover 1949). Gould’s position on these issues was further elaborated by Barry Supple (1959). In his important contribution to this topic, Supple emphasized the existence of a ‘‘drain of silver’’ from England, especially during the 1620s. Its principal cause, in turn, was the monetary chaos and currency debasement on the Continent that followed in the wake of the Thirty Years War. The upshot was a severe drop in demand, especially for British cloth. This was the phenomenon, according to Supple, that underlay ‘‘so many contemporary complaints concerning a scarcity of money’’ (1957:251). The heated argument among Mun, Miselden, and Malynes thus hinged on how to explain this deflationary process and monetary shortage. Both Gould and Supple tended to favor Malynes’s conclusion that the root of the problem lay in international currency manipulations. Since the pivotal issue was monetary problems, Malynes’s position appeared to be more ‘‘realistic’’ than that of Mun and Misselden, who saw the shortage of British coin as a secondary result of a negative trade balance. The key point, however, Supple maintained was that ‘‘the reiterated claims of these years that England had an unfavourable balance of trade were founded on uncomfortable fact.’’ From this, he immediately drew the conclusion that ‘‘much of the economic literature which historians have interpreted as ‘typical’ of mercantilism is, in fact, the product of a specific situation and a short-run crisis’’ (Supple 1957:251; see also Supple 1959:226f). Against this backdrop, Supple shared Judges’s, Coleman’s, and Schumpeter’s negative response to claims that mercantilism had systemlike characteristics. ‘‘In calling such writers ‘mercantilist’ there is the danger of implicitly attributing to them a continuity of doctrine, based on a set of supposedly logical principles, which was not theirs,’’ he warns (Supple 1959:228). The fact that the mercantilist writings were ‘‘consistent and pragmatic responses to consistent fluctuations in an economic environment whose basic elements were slow to change’’ hardly justifies its ‘‘treatment as a full-blown system.’’ Undoubtedly this argument might downgrade mercantilism’s claim to be a coherent system or ‘‘school’’ of economic thought. A mere ‘‘pragmatic response’’ to economic ‘‘reality,’’ however, does not preclude a process of interpretation or the use of concepts and even, crude as they might be,
242
Mercantilism
theories. Consequently, supposing that seventeenth-century authors of economic texts were nontheoretical or devoid of abstract thinking or ideological perspectives is unlikely to be a fruitful point of departure (Magnusson 1994a:1–20). A Current Perspective on Heckscher’s Mercantilism Certainly some of the criticisms that were unleashed by the publication of Heckscher’s work remain valid. The accumulation of subsequent economic-historical studies have provided a more realistic account both of economic policy making and of trade relations and monetary flows in Western Europe during the early modern period. In particular, few current scholars would accept the close relationship between economic ideas (however faulty) and the practice of economic policy that Heckscher argued existed during that historical epoch. Moreover, new theories concerning public policy formation, such as public choice based rent-seeking, have inspired new interpretations of mercantilism. Finally, increased knowledge of doctrinal history has forced a modification of Heckscher’s analysis of the contemporary literature. Most especially, the supposed strict demarcation line between liberal and mercantilist ideas, as well as Smith’s complete ‘‘otherness’’ vis-a`-vis his older British ‘‘mercantilist’’ colleagues, has been undermined. Clearly, there was convergence as well as deviation between earlier writers, such as Child, Barbon, Davenant, and the ‘‘Smithians’’ (Magnusson 2004). Against this backdrop, it is certainly debatable whether or not Heckscher’s Mercantilism has anything to offer today. Is the book worth reading for other than antiquarian reasons? I believe it is, for three reasons that I will discuss in order. First, I believe that it is still valid to speak of what might be called ‘‘mercantilist’’ economic ‘‘thinking’’ (for want of a better word). Surely, there are themes and systematic ideas in the economic literature between the early seventeenth century and Adam Smith that can legitimately be labeled ‘‘mercantilism.’’ British authors from Mun, Misselden, and Roger Cooke in the seventeenth century to Theodore Janssen and James Stuart in the eighteenth century, certainly display some degree of coherence in terms of shared ideas, concepts, and formulations on the role of trade, the nature of economic growth, the crucial importance of supply and demand for prices, and the role of the dirigist state. This makes it useful to speak, at very least, of a mercantilist ‘‘quasi-system’’ (Schumpeter 1968:194) of economic discourse and thinking that flourished in seventeenth- and eighteenth-century Britain.
Eli Heckscher and His Mercantilism Today
243
By the same token, however, it is extremely difficult to accept Schumpeter’s assertion that the mercantilists (or ‘‘consultant administrators’’ as he preferred to call them in recognition of their practical role and attitude) did no theorizing whatsoever. Of course, they obviously did not use the theories developed by twentieth-century economists. It is also possible to agree with Schumpeter that they were poor theorists. But is that really all there is to the matter? It seems unlikely that even Schumpeter believed that the mercantilists did not think systematically, however faulty, at all (Schumpeter 1968:143–160). It seems more fruitful to posit that writers such as Mun, Misselden, Child, Barbon, Law, and Davenant struggled with a common set of questions: how can a nation become rich; what, in fact, constitutes national wealth; what is the importance of money; and what is the role of market forces in setting interest rates and prices? In so doing, they used a common conceptual vocabulary whose meaning they also discussed. They argued among themselves, both implicitly and explicitly. Together they ventured forth to discover answers to the new problems raised by the functioning of a market economy. Thus, Heckscher’s conclusion that economic ideas in the seventeenth and eighteenth centuries shared at least some systemlike characteristics and cannot be regarded as just a type of ad hoc reasoning intended to solve particular practical problems, still seems valid. Second, the principal contribution of Heckscher’s treatment of the subject is not that he got everything right, but rather his impressive knowledge of the ideas and practices of the mercantilist era. Following Heckscher, perhaps in opposition to Viner as well as to Eklund and Tollison, I would argue that a skilled historical treatment of past economic ideas—including ‘‘mercantilism’’—has a number of important advantages over any stylized version of history. All too often, the latter are principally intended to confirm some preconceived theoretical proposition. Only well executed historical research permits the drawing of useful conclusion—in this case, concerning mercantilism—that have implications for the modern interpretation and application of economics. The most important lesson to be learned, however, is how earlier writers, such as those of the seventeenth century, came to grips with the problems that still confront us: how markets work, how institutions influence economic behavior, and the importance of uncertainty in the real economic world. An appreciation of their efforts might provide modern economists with a suitable dose of humility, as well as a realization that the last word in these centuries-old debates has not been said.
244
Mercantilism
Third, even though it has been severely criticized, Heckscher’s discussion of the interrelationship between economic ideas and policies remains highly stimulating. Undoubtedly Heckscher overstated both the ambition and the ability of early modern states to pursue selfinterested, coherent economic policies. They were usually too soft, too weak, and indeed too corrupt to succeed in such an endeavor. Nevertheless, Heckscher’s insistence that economic policy making also must be placed in an intellectual context is certainly appropriate. Statesmen of the period, in addition to pursuing their private interests, also were motivated by ideas of how the economy operated and the world works and, perhaps, even by higher religious or moral ideals. As Coats has remarked, in principle it can not be illegitimate to generalize also about economic policy. Thus it is clear that ‘‘economic policies’’—that myriad of decisions on various levels made by legislative or executive bodies of the state or subnational governments—must be guided by some visions and views concerning both how the ‘‘economy’’ operates and what are the ultimate ends of those policies. As Coats observed, ‘‘historians of economics recognise that without some conception of the way the economic system works, and the relationship between ‘means’ and ‘ends’, there can be no coherent policy making whatever.’’ It is, of course, always possible to argue that such coherent policy making was entirely lacking during the eighteenth and early nineteenth centuries. Such a position, however, is hardly realistic. Surely it is more fruitful to posit that even at that time administrators and politicians were at least searching for a set of means that would satisfy the ends that they had set for themselves. That, of course, is not to say that they were successful in that quest. At the same time, however, many of the interventions especially from the economic historians display, to a different degree, a reductionist tendency to immediately ‘‘explain’’ texts as the result of economical, political, and social circumstances. However, as we have argued, complex ‘‘circumstances’’ of the kind we deal with here are never immediately observable. They are always interpreted against the backdrop of a certain language, in concepts and words that are historically inherited and made intelligible to the actors. Once again we must stress that it is hardly enough to argue that Mun and Malynes were involved in a discussion that concerned merely pragmatic issues. Instead, it is totally clear that when they discussed ‘‘practical’’ economic matters such as the effect of monetary speculation on the trade balance, they did so with the help of a conceptual framework deeply rooted in
Eli Heckscher and His Mercantilism Today
245
earlier conceptions and views how the economy functioned and ought to be understood. Acknowledgments I want to thank especially Bob Coats for his valuable comments to an earlier draft of this essay. References Bloch, M. 1934. ‘‘Le mercantilisme, un e´tat d’e´sprit.’’ Annales 6:160–163. Coats, A. W. 1985. ‘‘Mercantilism, Yet Again!’’ In P. Roggi, ed., Gli economisti e la politica economica, 66–88. Naples: Edizione Scientifiche Italiane. Cole, C. W. 1950. ‘‘The Heavy Hand of Hegel.’’ Nationalism and Internationalism. New York: Richard R. Smith. Coleman, D. C. 1969. ‘‘Eli Heckscher and the Idea of Mercantilism.’’ In D. C. Coleman, ed., Revisions in Mercantilism, 92–117. London: Methuen & Co. ———. 1980. ‘‘Mercantilism Revisited.’’ Historical Journal 23, no. 4: 773. De Roover, R. 1949. Gresham on Foreign Exchange. Cambridge, MA: Harvard University Press. Earle, E. M. ed. 1950. Nationalism and Internationalism. New York: Columbia University Press. Ekelund, R. B., and R. D. Tollisson. 1997. Politicized Economies. College Station: Texas A&M University Press. Gould, J. D. 1954. ‘‘The Trade Depression of the Early 1620s.’’ Economic History Review 7:81–90. ———. 1955. ‘‘The Trade Crisis of the Early 1620s and English Economic Thought.’’ Journal of Economic History 15:121–133. Heaton, H. 1937. ‘‘Heckscher on Mercantilism.’’ Journal of Political Economy 14, no. 3: 370– 393. Heckscher, E. F. 1942. ‘‘Den ekonomiska historiens aspekter.’’ In Ekonomisk-historiska studier, 9–69. Stockholm: Bonniers. ———. 1950. ‘‘Multilateralism, Baltic Trade and the Mercantilists.’’ Economic History Review 2:219–228. ———. 1955. Mercantilism. 2 vols. London: Allen & Unwin. Judges, A. V. 1939. ‘‘The idea of a mercantile state.’’ In D. C. Coleman, ed., Revisions in Mercantilism. London: Methuen (1969), 35–60. Magnusson, L. 1994a. Mercantilism: the Shaping of an Economic Language. London: Routledge.
246
Mercantilism
———. 1994b. ‘‘Introduction.’’ In E. F. Heckscher, Mercantilism, xi–xxxv. London: Routledge. ———. 2004. The Tradition of Free Trade. London: Routledge. Price, J. M. 1961. ‘‘Multilateralism and/or Bilateralism: the Settlement of British Trade Balances with the North, c 1700.’’ Economic History Review 14:254–274. Rothbard, M. 1995. Economic Thought Before Adam Smith. Aldershot: Edward Elgar. Schmoller, G. 1896. The Mercantile System and its Historical Significance. New York: Macmillan. Schumpeter, J. A. 1968. History of Economic Analysis. London: George Allen & Unwin. Supple, B. 1957. ‘‘Currency and Commerce in the Early Seventeenth Century.’’ Economic History Review 10:239–255. ———. 1959. Commercial Crisis and Change in England 1600–1642. Cambridge: Cambridge University Press. Viner, J. 1930. ‘‘Early English Theories of Trade.’’ Journal of Political Economy 38. ———. 1935. Review of Heckscher’s Mercantilism. Economic History Review 6:99–101. ———. 1948. ‘‘Power versus Plenty as Objectives of Foreign Policy in the Seventeenth and Eighteenth Century.’’ World Politics 1:1–29. Wilson, C. 1949. ‘‘Treasure and Trade Balances: The Mercantilist Problem’’. Economic History Review 2:152–157. ———. 1951–52. ‘‘Treasure and Trade Balances: Further Evidence.’’ Economic History Review 4:231–242. ———. 1957. Profit and Power. Cambridge: Cambridge University Press. ———. 1969. Economic History and the Historians. London: Weidenfeld and Nicolson.
10
Mercantilism: Power and Plenty Through the Lens of Strategic Trade Policy Douglas A. Irwin
Ever since Adam Smith’s vigorous attack on ‘‘the mercantile system,’’ scholars have proposed and debated alternative ways of interpreting mercantilist economic thought and policy. Eli Heckscher’s two-volume Mercantilism (1935) stands out as a landmark contribution to this endeavor and continues to influence the way scholars view the seventeenth and eighteenth centuries. But as is true with any work of enduring interest, Heckscher also stimulated an ongoing debate over his interpretations of the period. Among the most important of the many reviews of Mercantilism in the leading economics journals was one by the formidable Jacob Viner. Viner (1935) clearly admired Heckscher’s work, especially as it related to the history of economic policy, and called it ‘‘an absolutely indispensable guide.’’ But he was less convinced that Heckscher had clearly stated or had accurately characterized English mercantilist thought on the relationship between national power and economic wealth. Far from subordinating factors such as wealth to considerations of power, as he understood Heckscher to say, Viner insisted that both power and wealth were desired goals and were viewed as complementary and not competing ends of mercantilist policy.1 Heckscher (1936) soon conceded that both ‘‘power’’ and ‘‘opulence’’ were always important goals of mercantilist economic policy. He reiterated, however, that greater significance was attached to power than to wealth, a ranking that was responsible for the ‘‘incessant commercial rivalries of the seventeenth and eighteenth centuries, which degenerated into military conflict.’’ With the subsequent rise of laissez faire, Heckscher argued, power gave way to wealth as the chief goal of economic policy. In a letter replying to Heckscher, Viner did ‘‘not deny that power was a more prominent objective of national policy in the 17 th and 18 th
248
Mercantilism
centuries than in the 19 th . . . . All I dispute . . . is your argument that the mercantilists regarded power and prosperity as competing alternatives, as between which choice had to be made.’’ With few exceptions, Viner insisted, ‘‘the English mercantilist literature treats power and prosperity as mutually supporting objectives . . . both were important objectives in their own right.’’2 This exchange with Heckscher prompted Viner to write his famous 1948 essay ‘‘Power versus Plenty as Objectives of Foreign Policy in the Seventeenth and Eighteenth Centuries.’’ In this article, Viner amassed an impressive array of textual evidence to support his contention that mercantilists believed that both wealth and power were ends to national policy and that each abetted the other. Viner’s exposition of mercantilist objectives proved convincing. The chapter entitled ‘‘Mercantilism as a System of Power’’ in the revised edition of Heckscher’s book was changed considerably in light of Viner’s article. In an addendum to the revised edition of Mercantilism, Heckscher (1955:359) conceded that Viner’s ‘‘evidence impresses me as being sufficiently strong to make me abandon my original thesis on the issue . . . . I have come to the conclusion that the difference between the mercantilist position and that which succeeded it was a difference of degree and not a difference of kind,’’ although Heckscher believed the difference to be a significant one. This paper revisits the debate about the relationship between power and plenty during the mercantilist period by proposing that mercantilism be viewed through the lens of strategic trade policy. Strategic trade policy, it is argued, provides a structure in which we can better understand some of the long-distance trade policies pursued by the major countries in the seventeenth century. In addition, this economic framework shows how power and plenty were indeed compatible, as Viner contended. The essential idea here is that, in many instances during the seventeenth century in particular, very few European firms competed over the same, potentially lucrative, long-distance trade routes. This competition among few players created conditions which were ripe for strategic behavior on the part of firms and governments. Under these circumstances, this trade competition was akin to a zero-sum game for the participants, and therefore state power could be exercised to the advantage of favored firms. Government promotion of a country’s firms could have abetted both the power and the plenty of that country
Mercantilism: Power and Plenty
249
at the expense of its trading rivals.3 The theory of strategic trade policy provides a framework in which we can better understand the relationship of government support for foreign trade monopolies. The term ‘‘mercantilism’’ encompasses a wide range of doctrines and policies and refers to a period that exhibits much diversity in terms of history and ideas and policy. The multifaceted objectives of state policy in the seventeenth and eighteenth centuries cannot be reduced into a single idea. Therefore, the spirit in which this paper is offered is simply that this additional way of thinking about mercantilism may be useful—perhaps appropriate only for a narrow range of phenomena, in particular, state-chartered foreign trade monopolies—and can possibly supplement other interpretive approaches.4 This essay begins by describing the basic framework behind strategic trade policy, and then moves on to describe briefly how English mercantilist thought is similar in outlook to the perspective one gains from strategic trade policy. The essay considers at greater length the East India trade, as contested by the English and Dutch East India Companies in the first few decades of the seventeenth centuries. As K. N. Chaudhuri (1978:20) notes, ‘‘If there was a perfect example of what we today understand as the spirit of mercantilism, the East India Company embodied it in its policy of harnessing political power and privileges to commercial purpose.’’ The economic characteristics of the East India trade, it turns out, map incredibly well into the assumptions underlying the canonical Brander and Spencer (1985) model of duopolistic trade competition. One of the strategic advantages of the Dutch over the English in this trade, it is suggested, is a subtle difference in the managerial compensation of the Dutch versus the English managers. A conclusion speculates on how far this analogy between strategic trade policy and mercantilism can work, and notes some of the qualifications to the analogy. Strategic Trade Policy Strategic trade policy arose as a part of the ‘‘new trade theory’’ in the 1980s which began exploring models of imperfect competition as applied to international trade.5 A strategic situation is one in which a small number of firms make interdependent decisions. Strategic trade policy examines the implications for commercial policy when the actions of each firm or government affect the profits and revenues of all
250
Mercantilism
other competitors. In the standard approach of perfect competition, there is little scope for firms to engage in strategic behavior because prices are taken as given. One of the leading, one might say canonical, models of strategic trade policy is the 1985 Journal of International Economics article by James Brander and Barbara Spencer entitled ‘‘Export Subsidies and International Market Share Rivalry.’’ This paper examined trade policy in the context of a very simple Cournot duopoly model and yet succeeded in generating much controversy and a large follow-on literature. The basic idea can be described as follows. Suppose two firms from different countries produce a homogenous good that they sell by competing in a third market. In this duopoly situation, national welfare can be represented by the profits of the country’s firm (there being no domestic consumption and hence no domestic consumer surplus to worry about). If the firms have identical costs and maximize profits by choosing the quantity of output to produce, then the Nash equilibrium is symmetric: each firm produces the same quantity and earns the same profit. Under certain conditions in this framework, a government export subsidy enables one (say, the domestic) firm to commit itself to a higher level of output than without the subsidy, forcing the output of the other (foreign) firm to contract. This subsidy increases the profits of the domestic firm at expense of the foreign firm, thus shifting the international distribution of profits to the home country. In addition, Brander and Spencer actually show that the optimal subsidy can increase national welfare because the gain in profits to the domestic firm exceeds the cost of the government subsidy. Of course, if both governments undertake such subsidies, the resulting Nash equilibrium is inefficient in that the welfare of all countries could be higher in the absence of such subsidies. Figure 10.1 illustrates firm behavior in this Cournot duopoly. The figure plots a schedule of the profit-maximizing level of output of each firm, taking as given various levels of output of the other firm. Notice that the higher is the level of foreign output, which drives down the market price, the lower is the profit-maximizing output of the domestic firm. The intersection of these ‘‘reaction functions’’ is the Nash equilibrium and determines the output levels of the two firms. In this setup, anything that the domestic firm can do to convince its foreign rival that it will produce more output will induce the foreign firm to reduce its output. The domestic firm will try to do this itself, but there are
Mercantilism: Power and Plenty
251
Figure 10.1 The Brander and Spencer (1985) model of strategic trade policy
limits to credible threats it can make. At this point, there is a role for government. As Grossman and Richardson (1985:11) point out, ‘‘in equilibrium, however, all such actions that are in the firm’s interest have already been taken, and threats of further thrusts by one firm are dismissed by other firms as mere bluffs. The government, by contrast, may have the ability to threaten and credibly precommit even after the firms attain oligopolistic equilibrium, shifting the equilibrium to obtain a nationally desirable distribution of profits.’’ Thus, a credible government subsidy can shift the domestic firm’s reaction function to the right, as illustrated in figure 10.1, and thus increase domestic output and lower foreign output. This ‘‘Stackelberg leader’’ equilibrium results in higher domestic profits and lower foreign profits. (Of course, the collusion equilibrium can increase the profits of both firms as they reduce output together.) Also shown on figure 10.1 are the associated ‘‘iso-profit’’ curves for the domestic firm—the combination of domestic and foreign output that leave domestic profits the same. The Stackelberg equilibrium, which involves higher domestic output and lower foreign output than the Nash equilibrium, is associated with higher profits for the domestic firm.
252
Mercantilism
While this model provides just an example of a way in which a government export subsidy can be beneficial and is subject to many theoretical and practical qualifications, the entire literature on strategic trade policy focuses on precisely those situations in which strategic interdependence creates opportunities for welfare improving government interventions. Mercantilist Economic Thought There is a remarkable similarity between English mercantilist economic thought on trade and the thinking that underlies theories of strategic trade policy.6 Both suggest that rents arising from imperfect competition are a prominent feature of international trade, both focus on capturing the gains from exporting for one’s own country at the expense of others by displacing rivals from the market, and both imply that an activist government can assist domestic firms engaged in international competition to the benefit of national welfare. A notable element of the mercantilist analysis was the perception that the gains from trade were derived from exporting domestic production or shipping goods between foreign markets. ‘‘Exportation is gain, but all Commodities Imported is loss,’’ stated Carew Reynell (1695:12). William Petty (1680:23) maintained that ‘‘The National Gain, by Foreign Trade, consist[s] either in vending home Commodities to Foreigners, or in Trading from Port to Port.’’ These gains from exports and the carrying trade, wholly unrelated to monetary concerns about the balance of trade, arose from the greater employment of labor, the larger mercantile profit, and the enhanced power and prestige of the nation which such commerce generated in increasing a country’s domestic production and improving its maritime capability. Distant overseas trade received a disproportionate amount of attention from the mercantilists, as these trades were associated with extraordinary prestige and profit. Yet mercantilists believed that increasing a country’s trade to be a difficult task because they took the total volume of world commerce to be fixed. The idea that ‘‘there is but a certain proportion of Trade in the World’’ led easily to William Petty’s conclusion (1690:82) that ‘‘the Wealth of every Nation consist[s] chiefly in the share which they have in the Foreign Trade with the whole Commercial World.’’ A fixed volume of trade meant that the fixed gains from trade had to be distributed among the trading countries of the world. Consequently, in-
Mercantilism: Power and Plenty
253
ternational trade took on the characteristics of a zero-sum game. Trade was set along certain ‘‘channels’’ that could not accommodate more traffic, hence entry was only possible by displacing existing merchants. This led to jealousy in trade and to the belief that one’s own country could be made better off by making others worse off. As Josiah Child (1693:160) argued, trade should be managed to ensure ‘‘that other Nations who are in competition with us for the same, may not wrest it from us, but that ours may continue and increase, to the diminution of theirs.’’ Child (1693:xxvi–xxvii) added that ‘‘no Trades deserve so much care to procure, and preserve, and encouragement to prosecute, as those that employ the most Shipping, although the Commodities transported be of small value in themselves; For, first, they are certainly the most profitable; for besides the gain accruing by the Goods, the Freight, which is in such Trades often more then the value of the Goods, is all profit to the Nation; besides, they bring with them a great access to Power.’’ The mercantilists concluded that activist trade policies in the form of export promotion schemes were essential to preserve and enhance national welfare. Widely advocated policies to increase exports included elimination of all government hindrances and disincentives to export, particularly export taxes; the enactment of direct financial support for exports, such as subsidies and drawbacks (tariff rebates on re-exported goods); and the implementation of indirect measures that would lower rates of interest and reduce wages. While mercantilists criticized importation of almost anything other than raw materials, the frequent association of the doctrine with high tariffs and import protection can be somewhat misleading because export promotion appears to have been the principal foreign trade goal of the mercantilists. Indeed, many writers recognized that import protection could adversely affect export volume. As Petyt (1680:61–62) explained, ‘‘for the opening of a sufficient Foreign Vent and Market for our Home Commodities . . . it is not only necessary to remove all unequal cloggs on mere Exportations, but also those on Imported Goods; because . . . the value of our English Exportation must be in a manner confined to the value of the Goods Imported.’’7 Yet competition over existing export markets was sure to be fierce if every government tried by export promotion to ensure that an ever larger share of world trade was captured by one’s own country, even if demand in these markets was not perfectly inelastic. Indeed, seventeenth-century competition in international trade fostered
254
Mercantilism
commercial rivalries that extended beyond the marketplace, even spilling over into military conflict as the many commercial wars of the period attest. This was decidedly not the era of doux commerce: Josiah Child again captured the temper of the times in writing, ‘‘All trade [is] a kind of warfare.’’8 Trade was not conceived as necessarily being mutually beneficial and the focus in contemporary writings was exclusively on national gain. Because of the narrow focus on national gain and the jealousy of other’s gains, there does not appear to be any recognition in the mercantilist literature of a prisoner’s dilemma element to government intervention, in which international cooperation to control export promotion could make all parties better off. The focus on export competition and rivalry makes mercantilist thought similar in outlook to that of strategic trade policy. Mercantilist Economic Policy: The East India Trade as an Illustration Having described the similarities of outlook in strategic trade policy and mercantilist economic policy, how might mercantilist trade rivalries may be amenable to interpretation by use of strategic trade models? To be specific, the Anglo-Dutch rivalry for the East India trade provides an example in of a mercantilist trade rivalry in which the economic conditions of that trade fit very well with the Brander and Spencer (1985) modeling framework.9 For many centuries before the foundation of the English East India Company, goods from India and Southeast Asia, particularly spices and silks, were in great demand in western Europe. Ancient and medieval trade between the two regions entailed the transportation of goods across the Asian continent in large caravans. Despite the exorbitant cost of transport, merchants still found it profitable to carry on a small trade with regularity. In 1498, the Portuguese explorer Vasco da Gama opened an entirely new route between Europe and Asia, traveling by sea around the Cape of Good Hope. Although this heralded a new age of trade between the two regions, a century elapsed before the sea route was fully exploited for commercial purposes.10 After individual English voyages to Asia in the 1590s yielded mixed results, a group of merchants founded the East India Company in 1600 as a joint-stock company designed to take advantage of the new trading opportunities with Asia. A royal charter from Queen Elizabeth I
Mercantilism: Power and Plenty
255
granted the company a fifteen-year exclusive monopoly to all trade beyond the Cape of Good Hope, as well as customs concessions and permission to export specie. These privileges were renewed and expanded by subsequent royal decrees. The purpose of the joint-stock arrangement was to allow investors to pool their capital, lease or purchase ships, hire crews and finance their provisions, and send the ships to India and Southeast Asia with bullion to make purchases and English goods to trade. Good fortune would have the ships return with a tremendous booty of Asian goods—such as pepper, cloves, nutmeg, indigo, silk, tea, and cotton goods—ready to fetch high prices in England and Europe and thereby compensate the joint stockholders several times over for their expense and risk. In the first half of the seventeenth century, the company was simply a shipping concern, arbitraging large price differentials between European and Asian markets, but not engaging in production. During its first decade, the English East India Company dispatched one ship a year on average to Asia, but by the mid-1620s sent about four ships a year (Steensgaard 1974:170). Losses due to shipwreck diminished with time—about 7 percent of the roughly 135 voyages before 1630 never returned (Krishna 1924:334ff)—although the threat of looting and piracy was an additional concern. Yet the East India trade proved to be profitable: the first two voyages earned a 95 percent profit, and net returns on early individual voyages ran as high as 230 percent (Chaudhuri 1965:209). These profits arose from the tremendous arbitrage opportunity open to the company: in the twenty years ending July 1620, purchases of £356,288 worth of goods in Asia fetched £1,914,600 in Europe (Khan 1923:17). This excludes transportation costs but is indicative of the markup (by a factor of five) achieved by the company. Pepper from Indonesia dominated the company’s trade in both value and volume for the first several decades of the East India trade. Profit margins shrank as the trade expanded, with pepper prices falling in Europe by roughly a quarter between 1609 and 1626 (Chaudhuri 1965:151; see also O’Rourke and Williamson 2002). For a few years before the formation of the English company, independent Dutch merchant groups had been engaged in routine commerce with Southeast Asia. In 1602, the States General (the Dutch governing body) initiated and helped finance the formation of the Dutch United East India Company, or VOC (Vereenigde Oostindische Compagnie), which was granted exclusive monopoly rights to engage in trade with Asia. By this stroke, Dutch trade was consolidated under
256
Mercantilism
the management of a single company in a government-sponsored effort to compete more effectively with trade rivals. Like the English, the Dutch mainly imported and reexported spices, with pepper accounting for nearly 60 percent of company trade by value in 1619–21 (Glamann 1958:13). But the VOC clearly dominated shipping volume in the early East India trade, returning sixty-five ships to England’s thirty-five during 1615–25 (Steensgaard 1974:170), with shipping losses comparable to those of the English (Bruijn, Gaastra, and Schoffer 1987, 1:91). How closely do the economic and institutional details of the earlyseventeenth-century East India trade (from 1600 to roughly 1630) bear resemblance to the Brander and Spencer (1985) analysis of duopolistic export competition? To assess whether the conditions of the East India trade conform to the assumptions of this framework, consider eight key elements of their model in light of the trade from the perspective of the English East India Company. Partial Equilibrium The partial equilibrium assumption is appropriate here because the early seventeenth-century East India trade was a very small and emerging trade, if a particularly intriguing and exotic one, in contrast with the more mundane intra-European trade that accounted for the overwhelming proportion of England’s international commerce. Even by 1663, according to Davis (1962:17), only 8,000 of the total 126,000 tonnage of ships engaged in England’s foreign trade was taken up servicing the East India trade route. Single Homogeneous Good Brander and Spencer abstract from product differentiation, an assumption in accord with the early East India trade in homogeneous commodities such as pepper and other spices. Duopoly with No Entry Brander and Spencer assume that only two firms of different nationality are engaged in export competition and that there is no free entry, despite the existence of monopoly profits. This assumption is quite accurate in describing the rivalry between the English and Dutch companies in this early period. Entry by other English and Dutch merchants was explicitly prohibited under the terms of the government charters
Mercantilism: Power and Plenty
257
granting the companies exclusive rights to the trade, rights legally enforceable against interlopers.11 Other European countries were not competitors at this time because their maritime capabilities were not sufficiently advanced in longdistance overseas trade. (France did not form an East India company until 1664; the Swedish East India Company dates from 1731.) Spain and Portugal ruled the seas in the sixteenth century, but different seas owing to a papal decree in 1493 that allocated trade with the Americas to Spain and trade with Asia to Portugal. In the early seventeenth century, however, Portugal distinguished itself only in the rapidity of its decline, a decline accelerated with forceful encouragement from the Dutch. The Portuguese quickly became a residual trader in Asia, and the English and Dutch accounted for about 80 percent of the East India pepper trade by the early 1620s (Chaudhuri 1965:144). Moreover, as Wake (1979) and Steensgaard (1974:171ff) document, overseas shipments of Asian goods to Europe after 1600 entirely displaced the more costly land transport of such goods via the Levant. Monopoly Profits Imperfect competition gives rise to monopoly profits that are shared by the duopoly. As previously discussed, there is little doubt that the early East India trade was lucrative, suggesting that such rents did exist for the trading companies engaged in the trade. An English East India Company investor received an annual average return of about 25 percent over the first decade of the trade (inclusive of shipping losses), about three times the market rate of interest in London, although profits fell off after 1615 owing in part to increased Dutch competition (Chaudhuri 1965:211–217). This was high enough to elicit numerous complaints against the monopoly from resentful merchants who were excluded from the trade. Figures in Krishna (1924:77) suggest comparable profitability for the VOC: a total dividend of 307.5 percent was returned to Dutch investors over 1605–20, amounting to roughly 20 percent annually. Cournot-Nash Game The assumptions of the Cournot duopoly framework, which entail two firms engaged in a static, noncooperative, one-period, simultaneousmove game, fit very few cases at any period in history but can be partially justified here. The competition was clearly noncooperative to
258
Mercantilism
judge by the tense relations between the firms and by the failure of collusive agreements to hold.12 The two companies made decisions about how many ships to return to Europe in a given year more or less simultaneous because of the nature of the annual sailing season to the East Indies. Ships returning to European ports had to depart within a window of less than six months to avoid the monsoon season in Asia and to avoid passage around the Cape of Good Hope in winter (Davis 1962:258).13 Neither firm had the ability to determine precisely how many rival ships were to be sent or returned in a given season. Once the separate, simultaneous decisions had been made, the annual season would end with all goods auctioned off on wholesale markets on arriving in European ports. Consequently, the firm’s choices are modeled as a repeated game, each of whose constituent subgames has a significant probability of being the terminal period (the monopoly charters could expire, be revoked, or be rendered moot by interlopers). The repeated one-shot game is motivated by the fact that, in the early years of the trade, the companies were busy arbitraging prices of goods between the two markets and were not incurring fixed costs of production, undertaking irreversible ship investments, or engaging in preemptive acquisition of territory. Under these conditions, the unique subgame perfect equilibrium may be described by the static Nash equilibrium in each period. Cournot (Quantity) Competition A key assumption of the Brander-Spencer model is that the firm’s choice variable is output (Cournot competition) instead of price (Bertrand competition) and that the firm makes no conjecture (a Cournot reaction) regarding the impact of changes in its output on its rival’s output. As Eaton and Grossman (1986) demonstrate, the optimal trade policy associated with the duopolistic rivalry depends critically on the nature of the competition and the conjectural variation entertained by each firm. In contrast to the Brander-Spencer finding that an export subsidy could increase national welfare under Cournot competition, Eaton and Grossman found that the optimal policy becomes an export tax with Bertrand competition. This is an important critique of the Brander-Spencer model because it is often difficult to assess whether firms are competing with quantities or prices. No such ambiguity arises in considering the East India trade because the firms clearly competed with quantities. The choice variable for both
Mercantilism: Power and Plenty
259
companies was the number of ships to have return in a given sailing season, thus determining the quantity of goods that would arrive at European ports in the coming months. Once both companies’ ships arrived from Asia loaded with a fixed quantity of goods, these goods would be auctioned on European wholesale markets. Because the cost of each voyage was paid up front and the private investors had an interest in receiving dividends immediately to retire current debts, in most instances all returning goods were placed onto the market on arrival. Less appealing is the idea that the companies necessarily entertained Cournot conjectures about how one’s competitor would respond to changes in one’s output. The assumption underlying the ‘‘Cournot conjecture’’ is that the domestic firm maximizes its profits by taking the foreign firm’s output as given, and that the foreign firm’s output will not change even if the domestic firm does change its output. This assumption is simply false because a change in the domestic firm’s output does change the profit-maximizing level of output by the foreign firm. Yet alternative assumptions about how a rival will respond to a change in one’s output are equally difficult to defend. In the case of ‘‘consistent conjectures,’’ in which a firm perfectly anticipates how its rival will react to a change in its output, there is no scope for strategic trade policies (Eaton and Grossman 1986). Indeed, the entire conjectural variation approach to these Cournot models is now out of favor with economists because of their arbitrary assumptions about the way in which firms interact with one another. Thus, in this critical aspect of these models, we must be agnostic for the moment. Constant Costs In contrast to many models of imperfect competition in international trade, Brander and Spencer assume that the single good can be produced under conditions of constant or increasing marginal cost and without recoverable fixed costs. The assumption of constant costs is taken to be a reasonable reflection of the two cost components of the East India trade, shipping and acquisition. First, the cost of ships was not fixed because there existed a well functioning capital market in ships. The East India Company could lease ships from the competitive intra-European shipping market in the event that it was short of available tonnage in a given year. Pricing in the rental market for ships was based not on a fixed charge per ship, but on a flat freight rate on the required tonnage.
260
Mercantilism
Second, the English and Dutch East India companies made only marginal demands on many goods produced and available in Southeast Asia during the early years of the trade. There is little evidence that the companies had much scope to exercise monopsonist power at this time. Prices in Asia could still vary from port to port depending on local conditions and from year to year depending on production. Yet cost prices of pepper in Asia were roughly constant in the early 1620s despite variation in the volume of English East India Company shipments (Chaudhuri 1965:148). Consequently, acquisition costs are treated as constant in a given year, with the various sources of supply in the Asian market as a whole ensuring that the trading companies could purchase as much as they could fill ships with at a given price. Exports to Third Markets Brander and Spencer assume that all trade occurs in third markets so that calculations of national economic welfare do not require an accounting of consumer surplus and the profits of the exporting firm become equivalently identified with national welfare. This assumption is reasonably accurate here: About 80–90 percent of English East India Company pepper was re-exported to northern Europe and the Mediterranean because of the limited market for pepper in England. Thus, the circumstances of the early East India trade seems to map well into the basic assumptions underlying the Brander-Spencer model. The next question is why the Dutch soon came to dominate the trade. Steensgaard (1965) finds that the Dutch had no advantage in shipping costs over the English in the East India trade. If the English and Dutch companies traded a homogeneous commodity for which they received the same price and paid the same acquisition and shipping costs, then what accounts for the larger trade volume and hence greater profits achieved by the Dutch? Linear demand and Cournot conjectures, for example, establish the presumption that the Nash equilibrium is perfectly symmetric in output and profits for identical firms. Neither firm had an entrenched advantage initially because there were no major sunk costs involved in the trade in the first quarter of the century, since forts and factories were to come later. While territorial control was an important long-run factor accounting for the Dutch success, at this stage all aspects of the East India trade were still open to each firm. So what was the difference between the firms?
Mercantilism: Power and Plenty
261
By all accounts, the competition between the two firms was intense, but the Dutch are considered to have been more aggressive and more ruthless than the English. While the rivalry between the English and Dutch companies in Asia was not plagued by outright commercial wars like the frequent conflicts in the European theater, the Dutch were tenacious in their efforts to eliminate foreign competitors from the spice islands of Indonesia. The VOC sought monopoly contracts with the principal spice-supplying regions to foreclose competitors and, unlike its English counterpart, was empowered to make treaties, acquire land, and build forts, thereby laying the groundwork for future colonization. The VOC penchant for looting rival vessels on occasion and intimidating English merchants in the region led to constant tensions and even an outbreak of hostilities in 1617–19 when the Dutch seized four English ships. After the mid-1620s, the English gradually ceded further trade with the Spice Islands to the Dutch and withdrew to trade in the far western points of Southeast Asia and with the Indian subcontinent at initially reduced trade volume and profits. While territorial control (either direct or indirect) ultimately provided the basis for the Dutch domination of the region’s international trade, a particular type of strategic policy facilitated the Dutch success. Comparison of the structure and objectives of each firm reveals a mechanism that facilitated Dutch ascendancy in the trade. This policy did not have anything to do with government subsidies, but with managerial incentives. The English East India Company was a private firm organized and run solely by merchants, with no government stake or involvement beyond granting the monopoly charter. It seems very clear from the institutional makeup of the company that its exclusive objective was to choose the quantity of pepper to ship each year to Europe in order to maximize the returns to investors. Thus the English East India Company is assumed to maximize profits, represented by the expression p ¼ pðx; x Þ x c x where p is profits, p is the inverse demand function for pepper in Europe, x is the quantity of pepper carried to Europe (and x by the Dutch), and c is the constant marginal (shipping and acquisition) cost. Choosing x to maximize profits yields the following first-order condition: qp=qx ¼ p c þ v x;
262
Mercantilism
where v ¼ ðdp=dxÞ ½1 þ ðdx =dxÞ and is the firm’s perceived marginal revenue, consisting of the first effect of changes in its output on price and the indirect effect of changes in its rivals output on price in response to changes in its own output (that is, the conjectural variation, with dp=dx ¼ dp=dx ). The institutional structure and economic objectives of the VOC, however, differed significantly from those of the English company. Before the formation of the VOC, Dutch trade with the East Indies was managed by the bewindhebbers, who made business decisions regarding the details of particular voyages and the sale of Asian goods in European markets. The bewindhebbers were directly accountable to shareholders (participanten), who were guided solely by the profit motive. But the granting of monopoly privileges and establishment of close government ties that accompanied the formation of the VOC in 1602 eroded the influence of the participanten on the bewindhebbers. In effect, stockholder control over the management of the company was supplanted by the government (Glamann 1958:6ff; Steensgaard 1974:126–141). In characterizing the objective function of the VOC, one must look to the particular incentives facing the bewindhebbers who determined the company’s shipping schedule. Steensgaard (1982:243) describes these incentives: Maximization of dividends was the obvious aim of the participant. The bewindhebbers were participants themselves, but for several reasons they would tend to have other aims. Their remuneration by provision made it their interest to maximize the turnover of the company, even at terms that were not advantageous to the participants. For the same reason they might prefer consolidation and maximum growth rather than dividends. The social and political distinction attached to their offices would work in the same direction. Finally, the close relations to the Dutch political leaders and the ultimate dependence on the political authorities for the continued existence of the company would tend to influence business decisions. So the charter of 1602, in spite of formal continuity, created a managerial group with interests deviating from those of the participants.’’ (Emphasis added)
As a consequence, for the first decades of the company, ‘‘the trading partnership was replaced by a permanent, anonymous capital; the bewindhebbers became a managerial group with close affiliations to the political authorities; the participants became holders of negotiable shares with not much more influence on company business than a holder of government bonds has on government policy, and the strategic aims of the company were radically changed’’ (Steensgaard 1982:250).
Mercantilism: Power and Plenty
263
By design, as mandated in the government’s monopoly charter, the bewindhebbers derived income both from their position as stockholders, for which they earned dividends that arose from profits, and from their role as managers, for which they earned a percentage of gross revenue. According to the charter, the sixty directors of the VOC were obligated to hold shares in the company and were to receive 1 percent of the value of equipment and ships and 1 percent of the proceeds of sales from each returning voyage. This changed in 1647, when the directors received a fixed salary as compensation (see Bruijn, Gaastra, and Schoffer 1987, 1:11ff). In essence, the bewindhebbers can be thought to have chosen x to maximize a linear combination of profits and revenue, L ¼ lfpðx; x Þ x c xg þ ð1 lÞf pðx; x Þg where l (which is between zero and one) is the weight put on profits in the objective function. This yields a first-order condition in which modifies the standard profit-maximizing condition in attaching to marginal cost the weight the firm places on profits in its objective function: qL=qx ¼ p l c þ v x : The key difference between this first-order condition and the one for the profit-maximizing English company is that here marginal costs are discounted by l ð
